CN108564423A - Malice occupy-place recognition methods, system, equipment and the storage medium of ticketing service order - Google Patents
Malice occupy-place recognition methods, system, equipment and the storage medium of ticketing service order Download PDFInfo
- Publication number
- CN108564423A CN108564423A CN201711499025.0A CN201711499025A CN108564423A CN 108564423 A CN108564423 A CN 108564423A CN 201711499025 A CN201711499025 A CN 201711499025A CN 108564423 A CN108564423 A CN 108564423A
- Authority
- CN
- China
- Prior art keywords
- occupy
- malice
- ticketing service
- service order
- place
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides the malice occupy-place recognition methods of ticketing service order, system, equipment and storage medium, method:The user information and behavioural information that ticketing service order is completed in database are obtained, therefrom extraction and the relevant characteristic variable of malice occupy-place;Calculate the value of information of each characteristic variable;The characteristic variable that the value of information is more than information threshold is filtered out, priori conditions are extracted;Regression model is trained using the characteristic variable filtered out;Real-time ticketing service order and its user information and behavioural information are obtained, corresponding priori conditions are seen if fall out, if then identification has malice occupy-place risk;The malice occupy-place probability of the corresponding lower single user of the real-time ticketing service order is calculated using trained regression model, and takes corresponding interception measure.The present invention identifies malice occupy-place risk by screening characteristic variable, using priori conditions, and trains the regression model for more accurately calculating malice occupy-place probability, improves the differentiation accuracy rate of malice occupy-place.
Description
Technical field
The present invention relates to the malice occupy-place recognition methods of Internet technical field more particularly to a kind of ticketing service order, system,
Equipment and storage medium.
Background technology
When normal ticketing service order is subscribed, such as during reservation, wound is singly filled in when page clicks next step and can be initiated
Booking request then obtains booking result.List is not paid in 15 minutes under user will close order.However, there are malicious users
It is subscribed in booking process but the not malice occupancy of pay invoice, causes corresponding losing one's seat for flight that can not sell, drop
The reservation experience of low normal users, while the flight of airline is sold and is caused damages.
In current air control system, empty occupy-place malicious user is identified using the artificial rule set based on business experience,
There are two main problems for the function.First, the setting of artificial rule is there are certain subjectivity, regular standard and threshold values
It is relatively low to set accuracy.Secondly, for malicious user by continuously attempting to easily identify and cracking air control rule, flexibility is poor.
It should be noted that information is only used for reinforcing the reason to the background of the disclosure disclosed in above-mentioned background technology part
Solution, therefore may include the information not constituted to the prior art known to persons of ordinary skill in the art.
Invention content
For the defects in the prior art, the problem to be solved in the present invention is, how to calculate the evil of ticketing service order in real time
Meaning occupy-place probability more accurately and comprehensively identifies malice occupy-place user.
According to an aspect of the present invention, a kind of malice occupy-place recognition methods of ticketing service order is provided, the method includes:
The user information and behavioural information that ticketing service order is completed in database are obtained, ticketing service order is completed described in part
It is marked as malice occupy-place, extraction and the relevant characteristic variable of malice occupy-place from the user information and the behavioural information;
It is not labeled in the value being completed in ticketing service order for being marked as malice occupy-place and based on each characteristic variable
For the difference for the value of malice occupy-place being completed in ticketing service order, the value of information of each characteristic variable is calculated;
The characteristic variable that the value of information is more than predetermined threshold value is filtered out, malice is being marked as based on the characteristic variable filtered out
The value of occupy-place being completed in ticketing service order extracts priori conditions;
Using characteristic variable training regression model Z=logit (p)=log (the odds)=β filtered out0+β1X1+β2X2+β3X3+…+βkXk, wherein p is the malice occupy-place probability of a ticketing service order, and odds is that the ticketing service order occurs malice occupy-place and do not send out
The ratio of raw malice occupy-place, odds=p/ (1-p), X1,X2,…,XkIt is right in the ticketing service order for each characteristic variable for filtering out
The value answered, β0For intercept, β1,β2,…,βkFor the corresponding regression coefficient of each characteristic variable filtered out;
Obtain real-time ticketing service order and its user information and behavioural information, judge the real-time ticketing service order user information and
Whether behavioural information meets priori conditions, if then judging that the real-time ticketing service order has malice occupy-place risk;
For the real-time ticketing service order with malice occupy-place risk, the user information based on the real-time ticketing service order and behavior
Information, the malice occupy-place probability of the real-time ticketing service order is calculated using trained regression model, and intercepts malice occupy-place probability
Real-time ticketing service order beyond probability threshold value.
Preferably, further include being adjusted to regression coefficient using following formula when the training regression model:
Wherein,For the lasso estimated values of regression coefficient;N is that ticketing service order is completed for train regression model
Total sample number, i takes from 1 to N;yiThe corresponding regression function of ticketing service order is completed for i-th,yi∈ { 0,1 }, yi=1 represents ticketing service order malice occupy-place, yi=0 represents the non-evil of ticketing service order
Meaning occupy-place;K is the characteristic variable number for training regression model filtered out, and j takes from 1 to k;For penalty term;t
For binding occurrence, t passes through the adaptively selected expectation estimation value minimum for making prediction error.
Preferably, the step of real-time ticketing service order for intercepting malice occupy-place probability beyond probability threshold value includes:It will dislike
Occupy-place probability of anticipating divides malice occupy-place grade according to multiple probability threshold values, to the real-time ticketing service order pair of different malice occupy-place grades
The lower single user answered takes corresponding interception measure respectively.
Preferably, the interception measure includes:The real-time ticketing service for being less than the first probability threshold value for malice occupy-place probability is ordered
Single corresponding lower single user, allows its ticket reservation behavior;First probability threshold value and small is more than or equal to for malice occupy-place probability
In the corresponding lower single user of real-time ticketing service order of the second probability threshold value, identifying code is popped up;Malice occupy-place probability is more than etc.
In the corresponding lower single user of real-time ticketing service order of the second probability threshold value, refuses its ticketing service and subscribe behavior.
Preferably, when carrying out priori conditions judgement, if a real-time ticketing service order is unsatisfactory for priori conditions, judge this in real time
Ticketing service order does not have malice occupy-place risk, and the ticketing service of the corresponding lower single user of the real-time ticketing service order is allowed to subscribe behavior.
Preferably, the step of obtaining real-time ticketing service order and its user information and behavioural information include:Obtain real-time ticketing service
Order, and obtain the user information and behavioural information of the characteristic variable filtered out described in real-time ticketing service order correspondence.
Preferably, further include becoming to the feature filtered out before using the characteristic variable training regression model filtered out
Amount carries out the step of data prediction.
According to another aspect of the present invention, a kind of malice occupy-place identifying system of ticketing service order is provided, the system comprises:
Sample acquisition module obtains and the user information and behavioural information of ticketing service order is completed in database, described in part
Ticketing service order is completed and is marked as malice occupy-place, extraction and malice occupy-place phase from the user information and the behavioural information
The characteristic variable of pass;
Variable processing module, based on each characteristic variable in the value being completed in ticketing service order for being marked as malice occupy-place
Difference in the value being completed in ticketing service order for being not labeled as malice occupy-place, calculates the value of information of each characteristic variable;
Priori processing module filters out the characteristic variable that the value of information is more than predetermined threshold value, based on the characteristic variable filtered out
In the value being completed in ticketing service order for being marked as malice occupy-place, priori conditions are extracted;
Model training module, using the characteristic variable training regression model Z=logit (p)=log (odds) filtered out=
β0+β1X1+β2X2+β3X3+…+βkXk, wherein p is the malice occupy-place probability of a ticketing service order, and odds is that the ticketing service order is disliked
Meaning occupy-place and the ratio that malice occupy-place does not occur, odds=p/ (1-p), X1,X2,…,XkIt is each characteristic variable for filtering out at this
Corresponding value, β in ticketing service order0For intercept, β1,β2,…,βkFor the corresponding regression coefficient of each characteristic variable filtered out;
Malice identification module obtains real-time ticketing service order and its user information and behavioural information, judges that the real-time ticketing service is ordered
Whether single user information and behavioural information meets priori conditions, if then judging that the real-time ticketing service order has malice occupy-place wind
Danger;
Probability evaluation entity, for the real-time ticketing service order with malice occupy-place risk, based on the real-time ticketing service order
User information and behavioural information, the malice occupy-place probability of the real-time ticketing service order is calculated using trained regression model, and is blocked
Cut the real-time ticketing service order that malice occupy-place probability exceeds probability threshold value.
According to another aspect of the present invention, a kind of malice occupy-place identification equipment of ticketing service order is provided, including:Processing
Device;And memory, the executable instruction for storing the processor;Wherein, the processor is configured to via execution institute
Executable instruction is stated to execute the step of malice occupy-place of above-mentioned ticketing service order identifies.
According to another aspect of the present invention, computer readable storage medium is provided, computer program is stored thereon with, it should
The step of malice occupy-place identification of above-mentioned ticketing service order is realized when program is executed by processor.
In view of this, the advantageous effect of the present invention compared with prior art is:
The present invention filters out the characteristic variable with high resolution from user information and behavioural information, and it is more accurate to establish
The regression model of the ticketing service order identification malice occupy-place of specification, can recognize that more malice occupy-place users, reduces to normal
The erroneous judgement of user reduces cost and improves air control efficiency.Meanwhile performance period the characteristics of for different times malice occupy-place user
Property update, dynamically control the risk brought of malicious user.
It should be understood that above general description and following detailed description is only exemplary and explanatory, not
The disclosure can be limited.
Description of the drawings
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the application
Example, and the principle together with specification for explaining the application.It should be evident that the accompanying drawings in the following description is only the disclosure
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 shows that a kind of the step of malice occupy-place recognition methods of ticketing service order in exemplary embodiment of the present is illustrated
Figure;
Fig. 2 shows a kind of sigmoid function curve diagrams in exemplary embodiment of the present;
Fig. 3 shows a kind of modular structure of the malice occupy-place identifying system of ticketing service order in exemplary embodiment of the present
Figure;
Fig. 4 shows a kind of schematic diagram of the malice occupy-place identification equipment of ticketing service order in exemplary embodiment of the present;
Fig. 5 shows a kind of schematic diagram of computer readable storage medium in exemplary embodiment of the present.
Specific implementation mode
Example embodiment is described more fully with reference to the drawings.However, example embodiment can be with a variety of shapes
Formula is implemented, and is not understood as limited to example set forth herein;On the contrary, thesing embodiments are provided so that the present invention will more
Fully and completely, and by the design of example embodiment comprehensively it is communicated to those skilled in the art.Described feature, knot
Structure or characteristic can be in any suitable manner incorporated in one or more embodiments.
In addition, attached drawing is only the schematic illustrations of the present invention, it is not necessarily drawn to scale.Identical attached drawing mark in figure
Note indicates same or similar part, thus will omit repetition thereof.Some block diagrams shown in attached drawing are work(
Energy entity, not necessarily must be corresponding with physically or logically independent entity.Software form may be used to realize these work(
Energy entity, or these functional entitys are realized in one or more hardware modules or integrated circuit, or at heterogeneous networks and/or place
These functional entitys are realized in reason device device and/or microcontroller device.
Fig. 1 shows the step schematic diagram of the malice occupy-place recognition methods of ticketing service order in embodiment.Shown in referring to Fig.1, this
The malice occupy-place recognition methods of ticketing service order includes in embodiment:
Step S101, the user information and behavioural information that ticketing service order is completed in database are obtained, ticket is partly completed
Business order is marked as malice occupy-place, extraction and the relevant characteristic variable of malice occupy-place from user information and behavioural information.
Wherein, user information is used to mark the user property of ticketing service order, such as user's gender, user ascription area etc.;
Behavioural information is used to mark the behavior property of ticketing service order, such as wound single time, creates single location etc..The value of characteristic attribute
The slightly difference according to the difference of qualitative variable and quantitative variable.Male is divided into for qualitative variable, such as user's gender, user
And women, male can be denoted as 1, women is denoted as 0.It is for user's gender this characteristic variable, value represents when being 1
Male, value represent women when being 0.For quantitative variable, such as year booking number, its actual value conduct can be directly used
Value.Such as user year booking 10 times, then its characteristic variable year booking number value is 10;Another user year booking
5 times, then its characteristic variable year booking number value is 5.
Step S102, based on each characteristic variable the value being completed in ticketing service order for being marked as malice occupy-place and
It is not labeled as the difference for the value of malice occupy-place being completed in ticketing service order, calculates the value of information of each characteristic variable.
In the follow-up model trained for calculating malice occupy-place probability, and not all characteristic variable is all as input number
According to, and to filter out the high characteristic variable of importance.Because the present embodiment will solve the problems, such as it is classification problem, that is, distinguish malice
Occupy-place and non-malicious occupy-place.If table of some characteristic variable in malice occupy-place ticketing service order and non-malicious occupy-place ticketing service order
Now different is exactly discrimination, and it is exactly the characteristic variable for having high discrimination that can better discriminate between open two class ticketing service orders.
It is mainly shown as in the ticketing service order of malice occupy-place and the ticketing service order of non-malicious occupy-place, area is distributed in the value of this feature variable
Not.
Step S103, the characteristic variable that the value of information is more than predetermined threshold value is filtered out, based on the characteristic variable filtered out in quilt
Labeled as the value of malice occupy-place being completed in ticketing service order, priori conditions are extracted.
For example, during the value of information of above-mentioned calculating characteristic variable, certain characteristic variables, such as air ticket behavior are found
Feature has good filtration for malice occupy-place user, therefore uses this feature variable as the priori rule before mode input
Then.Being represented by the ticketing service order of priori rules has malice occupy-place risk, therefore subsequently calculates its evil by regression model again
Meaning occupy-place probability, the ticketing service order for not meeting priori rules are automatically labeled as normal order.I.e. priori conditions be ticketing service order into
Enter the filtering before regression model, there is malice to account for possible ticketing service order using priori rules delineation first, reuse recurrence mould
Type calculating judges its malice occupy-place probability.
Step S104, using characteristic variable training regression model Z=logit (p)=log (the odds)=β filtered out0+β1X1+β2X2+β3X3+…+βkXk, wherein p is the malice occupy-place probability of a ticketing service order, and odds is that the ticketing service order occurs maliciously to account for
Position and the ratio that malice occupy-place does not occur, odds=p/ (1-p), X1,X2,…,XkIt is each characteristic variable for filtering out in the ticketing service
Corresponding value, β in order0For intercept, β1,β2,…,βkFor the corresponding regression coefficient of each characteristic variable filtered out.
Further include data prediction, including number specifically, before using the characteristic variable filtered out training regression model
According to cleaning, exceptional value, data conversion, Data Discretization, one-hot coding etc. are removed.Model training refers to machine learning, by
Correspondence between a part of input data having and output data generates a function, input is mapped to suitable defeated
Go out, such as classifies.In machine learning, the process that model parameter is found according to given data is exactly to train, and is finally searched
Mapping is referred to as training the model come.In the present embodiment, input data is the characteristic variable filtered out, and output data is malice
The classification of occupy-place/non-malicious occupy-place, the process for finding input data and the parameter of output data correspondence function is model
Trained process.
Specifically, it is assumed that a ticketing service order is that malice occupy-place is labeled as 1, and a ticketing service order is non-malicious occupy-place mark
It is denoted as 0, to solve the problems, such as being to predict whether the classification of a ticketing service order is malice occupy-place, can be write as yi∈{0,1}。yi
=1 represents ticketing service order malice occupy-place, yi=0 represents ticketing service order non-malicious occupy-place.Logistic regression is one and uses logic letter
Several regression process, logistic regression formula:Z=logit (p)=log (odds)=β0+β1X1+β2X2+β3X3+…+βkXk, wherein
β0For intercept, β1, β2..., βkFor regression coefficient, X1, X2..., XkIt is characterized variable, p is the general of ticketing service order malice occupy-place
Rate.The codomain of the ratio odds=p/ (1-p) that the probability of odds expression events event generation and do not occur, odds are 0 to just
Infinite, probability is bigger, and the possibility that event occurs and (malice occupy-place occurs) is bigger.If characteristic variable is multiplied by its regression coefficient
It is non-significant to be different from 0, illustrate that this feature variable does not contribute prediction output data (i.e. whether malice occupy-place) significantly, it will
It is removed from regression model.
It can be expressed as correspondingly, Probability p is class=1:
FunctionIt is referred to as sigmoid functions, is one and changes with z values and become in [0,1] section
The value range of the sigmoid function of change, z is positive and negative infinite, and corresponding function curve is as shown in Figure 2.According to business demand, adjustment is best
Threshold values, i.e., if p>Specific threshold is considered as the i.e. malice occupy-places of class=1.
Above-mentioned Logic Regression Models are based on largely existing malice occupy-place order and non-malicious occupy-place order and are used as training
Data identify the characteristic variable with high discrimination, and assign the rational weight of each characteristic variable (i.e. regression coefficient);Root
According to Logic Regression Models can be calculated in conjunction with real time data and historical data a ticketing service order whether the probability of malice occupy-place
Value.Preferably, by the way of Logic Regression Models combination Lasso models, the influence of data fluctuations can be reduced, model is made to have
There is preferable stabilization.
Herein, real time data and historical data all when for carrying out the required input data of model training.Historical data packet
Include two kinds:1) it is not changed according to time change that user, which is the characteristic variables such as masculinity femininity, this characteristic variable if there is
Historical data can be directly from data base call;2) if a certain characteristic variable is the data in past 1 year of statistics, if daily
Efficiency can be influenced by being calculated again when list under Real time request, that can first count well and go 364 days, only need to add daily to work as
Its data when lower single, would not influence the computational efficiency of system.Real time data is also classified into two classes:1) it is used for history
Data, which are done, splices;2) certain characteristic variables are exactly real-time data, that is, are used for describing certain performances of user's current point in time.
Specifically, Lasso (Least Absolute Shrinkage and Selection Operator) algorithm is one
Kind is carried out at the same time the regression analysis of feature variables selection and regularization (mathematics), it is intended to which the prediction for enhancing statistical model is accurate
Property and interpretation.It is to belong to Elastic Net families with the generalized linear model of parameter punishment, punish back that Lasso, which is returned,
Return the order of magnitude of coefficient.Its main thought is the summation of weight square to be added in loss function, therefore can stablize ginseng
Number is estimated to improve the accuracy of prediction, reduces variation degree and improves the precision of linear regression model (LRM).
Lagrangian Form can also be write, i.e.,:
Wherein,For the lasso estimated values of regression coefficient;N is that ticketing service order is completed for train regression model
Total sample number, i takes from 1 to N;yiThe corresponding regression function of ticketing service order is completed for i-th,yi∈ { 0,1 }, y)=1 represents ticketing service order malice occupy-place, yi=0 represents the non-evil of ticketing service order
Meaning occupy-place;K is the characteristic variable number for training regression model filtered out, and j takes from 1 to k;For penalty term;t
For binding occurrence, the regression coefficient of certain characteristic variables can be caused to be 0 if t values are sufficiently small;T is made pre- by adaptively selected
The expectation estimation value for surveying error is minimum.
The degree that Lasso returns complexity adjustment is controlled by parameter lambda, and the bigger linear models more to variable of λ is punished
Penalize dynamics bigger, to finally obtain the less model of a variable.The first row of object function returns mould with conventional linear
Type is identical, i.e., it is desirable that obtaining corresponding independent variable factor beta, practical dependent variable y and prediction dependent variable are minimized with this
βxBetween error sum of squares.And linear Elastic Net and the difference of linear regression are that whether there is or not these of the second row
A constraint, the independent variable coefficient that linear Elastic Net are intentionally got are within the scope of one controlled by t.This constraint
The reason of complexity adjustment can be carried out by being Elastic Net models, and Lasso recurrence can carry out Variable Selection and complexity adjustment.
Step S105, real-time ticketing service order and its user information and behavioural information are obtained, judges the real-time ticketing service order
Whether user information and behavioural information meet priori conditions, if then judging that the real-time ticketing service order has malice occupy-place risk;
Step S106, for the real-time ticketing service order with malice occupy-place risk, user information and row based on the real-time ticketing service order
For information, the malice occupy-place probability of the real-time ticketing service order is calculated using trained regression model, and it is general to intercept malice occupy-place
Rate exceeds the real-time ticketing service order of probability threshold value.
Specifically, obtaining real-time ticketing service order, and obtain the characteristic variable filtered out described in real-time ticketing service order correspondence
User information and behavioural information.When carrying out priori conditions judgement, if a real-time ticketing service order is unsatisfactory for priori conditions, judge
The real-time ticketing service order does not have malice occupy-place risk, and the ticketing service of the corresponding lower single user of the real-time ticketing service order is allowed to subscribe row
For.For meeting the real-time ticketing service order of priori conditions, show that there are malice occupy-place risks for its lower single user, therefore using recurrence
Model calculates its malice occupy-place probability.Wherein, the step of malice occupy-place probability exceeds the real-time ticketing service order of probability threshold value is intercepted
Including:Malice occupy-place probability is divided into malice occupy-place grade according to multiple probability threshold values, to the real-time of different malice occupy-place grades
The corresponding lower single user of ticketing service order, takes corresponding interception measure respectively.For example, it is general to be less than first for malice occupy-place probability
The corresponding lower single user of real-time ticketing service order of rate threshold value, allows its ticket reservation behavior;Malice occupy-place probability is more than etc.
In the first probability threshold value and less than the corresponding lower single user of real-time ticketing service order of the second probability threshold value, identifying code is popped up;For
Malice occupy-place probability is more than or equal to the corresponding lower single user of real-time ticketing service order of the second probability threshold value, refuses its ticketing service and subscribes row
For.
Further, daily record has recorded the variation of the characteristic variable of input model simultaneously, and to the value of characteristic variable point
Cloth carries out statistical monitoring, carrys out iteration more new model.For example, if male to female ratio is man in training data malice occupy-place sample
The distribution of 70% female 30% is male 55% female 45% in non-malicious occupy-place sample, but through the data on line after a period of time point
Cloth becomes male 45% female 55%, this when of data (characteristic variable) are just changed, and needs iteration update mould in time
Type.
To sum up, the present embodiment is by regression model, based on largely existing malice occupy-place order and non-malicious occupy-place order
As training data, the characteristic variable with high discrimination is identified, and assign each characteristic variable rational weight;According to mould
Type can be calculated in conjunction with real time data and historical data ticketing service order whether the probability value of malice occupy-place.Using logistic regression mould
The mode of type combination Lasso models, can reduce the influence of data fluctuations, and model is made to have preferable stability.And pass through
The data variation of the daily record data monitoring feature variable of log recording, is adjusted model, realizes dynamic update iteration.
The present invention also provides a kind of malice occupy-place identifying systems of ticketing service order, including:
Sample acquisition module obtains and the user information and behavioural information of ticketing service order is completed in database, described in part
Ticketing service order is completed and is marked as malice occupy-place, extraction and malice occupy-place phase from the user information and the behavioural information
The characteristic variable of pass;
Variable processing module, based on each characteristic variable in the value being completed in ticketing service order for being marked as malice occupy-place
Difference in the value being completed in ticketing service order for being not labeled as malice occupy-place, calculates the value of information of each characteristic variable;
Priori processing module filters out the characteristic variable that the value of information is more than predetermined threshold value, based on the characteristic variable filtered out
In the value being completed in ticketing service order for being marked as malice occupy-place, priori conditions are extracted;
Model training module, using the characteristic variable training regression model Z=logit (p)=log (odds) filtered out=
β0+β1X1+β2X2+β3X3+…+βkXk, wherein p is the malice occupy-place probability of a ticketing service order, and odds is that the ticketing service order is disliked
Meaning occupy-place and the ratio that malice occupy-place does not occur, odds=p/ (1-p), X1,X2,…,XkIt is each characteristic variable for filtering out at this
Corresponding value, β in ticketing service order0For intercept, β1,β2,…,βkFor the corresponding regression coefficient of each characteristic variable filtered out;
Malice identification module obtains real-time ticketing service order and its user information and behavioural information, judges that the real-time ticketing service is ordered
Whether single user information and behavioural information meets priori conditions, if then judging that the real-time ticketing service order has malice occupy-place wind
Danger;
Probability evaluation entity, for the real-time ticketing service order with malice occupy-place risk, based on the real-time ticketing service order
User information and behavioural information, the malice occupy-place probability of the real-time ticketing service order is calculated using trained regression model, and is blocked
Cut the real-time ticketing service order that malice occupy-place probability exceeds probability threshold value.
With reference to the function structure chart of the malice occupy-place identifying system of ticketing service order shown in Fig. 3.Wherein, above-mentioned sample obtains
Modulus block, variable processing module, priori processing module, model training module are respectively positioned in malice occupy-place identification model.Malice is known
Other module and probability evaluation entity are located in air control module.First, according to the sample data in database, characteristic variable is carried out
Screening and processing, that is, execute Feature Engineering, carries out machine learning (i.e. supervised learning) using treated characteristic variable, is formed
The model of malice occupy-place for identification exports the malice occupy-place identification model to air control module.
When user is when external (i.e. client is different from internal system) starts to subscribe ticketing service order, system acquisition is in fact
When data, and combine the historical data in database, data prediction, priori identification carried out to the characteristic variable of the ticketing service order
Deng operation, and input malice occupy-place Model Identification its malice occupy-place probability.According to the malice occupy-place probability of identification, graduation carries out
Air control is handled, and for malice occupy-place order, prevents it from continuing to make a reservation for by operations such as identifying code, interceptions;For non-malicious occupy-place
Order then allows it to continue intended flow.
Meanwhile the data variation of the real-time monitoring feature variable of system log, model is adjusted, realizes dynamic update
Iteration.Enable adaptation to the value distribution of the characteristic variables such as continually changing user property and behavior property.
In an exemplary embodiment of the present invention, a kind of malice occupy-place identification equipment of ticketing service order, the ticketing service are also provided
The malice occupy-place identification equipment of order may include processor, and the executable instruction for storing processor memory.
Wherein, processor is configured to execute the ticketing service order described in any one above-mentioned embodiment via executable instruction is executed
The step of malice occupy-place recognition methods.
Person of ordinary skill in the field it is understood that various aspects of the invention can be implemented as system, method or
Program product.Therefore, various aspects of the invention can be embodied in the following forms, i.e.,:It is complete hardware embodiment, complete
The embodiment combined in terms of full Software Implementation (including firmware, microcode etc.) or hardware and software, can unite here
Referred to as circuit, " module " or " system ".
The malice occupy-place identification equipment of the ticketing service order of this embodiment according to the present invention is described referring to Fig. 4
400.The malice occupy-place identification equipment 400 for the ticketing service order that Fig. 4 is shown is only an example, should not be to the embodiment of the present invention
Function and use scope bring any restrictions.
As shown in figure 4, the malice occupy-place identification equipment 400 of ticketing service order is showed in the form of universal computing device.Ticketing service
The component of the malice occupy-place identification equipment 400 of order can include but is not limited to:It is at least one processing unit 410, at least one
Storage unit 420, the bus 430 of connection different system component (including storage unit 420 and processing unit 410), display unit
440 etc..
Wherein, storage unit has program stored therein code, and program code can be executed by processing unit 410 so that processing is single
Member 410 execute described in the malice occupy-place recognition methods part of the above-mentioned ticketing service order of this specification according to the various examples of the present invention
The step of property embodiment.For example, processing unit 410 can execute step as shown in fig. 1.
Storage unit 420 may include the readable medium of volatile memory cell form, such as Random Access Storage Unit
(RAM) 4201 and/or cache memory unit 4202, it can further include read-only memory unit (ROM) 4203.
Storage unit 420 can also include program/utility with one group of (at least one) program module 4205
4204, such program module 4205 includes but not limited to:Operating system, one or more application program, other program moulds
Block and program data may include the realization of network environment in each or certain combination in these examples.
Bus 430 can be to indicate one or more in a few class bus structures, including storage unit bus or storage
Cell controller, peripheral bus, graphics acceleration port, processing unit use the arbitrary bus structures in a variety of bus structures
Local bus.
The malice occupy-place identification equipment 400 of ticketing service order can also with one or more external equipments 500 (such as keyboard,
Sensing equipment, bluetooth equipment etc.) communication, the malice occupy-place with the ticketing service order can be also enabled a user to one or more to be known
The equipment communication that other equipment 400 interacts, and/or with enable the ticketing service order malice occupy-place identification equipment 400 and one or
Any equipment (such as router, modem etc.) communication that a number of other computing devices are communicated.This communication can
To be carried out by input/output (I/O) interface 450.Also, the malice occupy-place identification equipment 400 of ticketing service order can also pass through
Network adapter 460 and one or more network (such as LAN (LAN), wide area network (WAN) and/or public network, such as
Internet) communication.Network adapter 460 can be communicated by bus 430 with other modules of electronic equipment 400.It should be understood that
Although not shown in the drawings, other hardware and/or software mould can be used in conjunction with the malice occupy-place identification equipment 400 of ticketing service order
Block, including but not limited to:Microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape
Driver and data backup storage system etc..
In an exemplary embodiment of the present invention, a kind of computer readable storage medium is additionally provided, meter is stored thereon with
The ticketing service order described in any one above-mentioned embodiment may be implemented in calculation machine program, the program when being executed by such as processor
The step of malice occupy-place recognition methods.In some possible embodiments, various aspects of the invention are also implemented as one
The form of kind program product comprising program code, when described program product is run on the terminal device, said program code
For make the terminal device execute the above-mentioned ticketing service order of this specification malice occupy-place recognition methods description according to this hair
The step of bright various illustrative embodiments.
Refering to what is shown in Fig. 5, describing the program product for realizing the above method according to the embodiment of the present invention
600, portable compact disc read only memory (CD-ROM) may be used and include program code, and can in terminal device,
Such as it is run on PC.However, the program product of the present invention is without being limited thereto, in this document, readable storage medium storing program for executing can be with
To be any include or the tangible medium of storage program, the program can be commanded execution system, device either device use or
It is in connection.
The arbitrary combination of one or more readable mediums may be used in described program product 600.Readable medium can be can
Read signal medium or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared
The system of line or semiconductor, device or device, or the arbitrary above combination.The more specific example of readable storage medium storing program for executing is (non-
Exhaustive list) include:Electrical connection, portable disc, hard disk, random access memory (RAM) with one or more conducting wires,
Read-only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, the read-only storage of portable compact disc
Device (CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
The computer readable storage medium may include the data letter propagated in a base band or as a carrier wave part
Number, wherein carrying readable program code.Diversified forms, including but not limited to electromagnetism may be used in the data-signal of this propagation
Signal, optical signal or above-mentioned any appropriate combination.Readable storage medium storing program for executing can also be any other than readable storage medium storing program for executing
Readable medium, which can send, propagate either transmission for being used by instruction execution system, device or device or
Person's program in connection.The program code for including on readable storage medium storing program for executing can transmit with any suitable medium, packet
Include but be not limited to wireless, wired, optical cable, RF etc. or above-mentioned any appropriate combination.
It can be write with any combination of one or more programming languages for executing the program that operates of the present invention
Code, described program design language include object oriented program language-Java, C++ etc., further include conventional
Procedural programming language-such as " C " language or similar programming language.Program code can be fully in user
It executes on computing device, partly execute on a user device, being executed as an independent software package, partly in user's calculating
Upper side point is executed or is executed in remote computing device or server completely on a remote computing.It is being related to far
In the situation of journey computing device, remote computing device can pass through the network of any kind, including LAN (LAN) or wide area network
(WAN), it is connected to user calculating equipment, or, it may be connected to external computing device (such as utilize ISP
To be connected by internet).
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can also be realized by software realization in such a way that software is in conjunction with necessary hardware.Therefore, according to the present invention
The technical solution of embodiment can be expressed in the form of software products, the software product can be stored in one it is non-volatile
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are so that a calculating
Equipment (can be personal computer, server or network equipment etc.) executes the above-mentioned ticketing service according to embodiment of the present invention
The malice occupy-place recognition methods of order.
To sum up, malice occupy-place recognition methods, system, equipment and the storage medium of ticketing service order of the invention, passes through recurrence
Model is identified and is distinguished with high based on largely existing malice occupy-place order and non-malicious occupy-place order as training data
The characteristic variable of degree, and assign each characteristic variable rational weight;Real time data and historical data can be combined according to model
Calculate ticketing service order whether the probability value of malice occupy-place.By the way of Logic Regression Models combination Lasso models, it can subtract
The influence of few data fluctuations, makes model have preferable stability.And become by the daily record data monitoring feature of log recording
The data variation of amount, is adjusted model, realizes dynamic update iteration.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the present invention
Its embodiment.This application is intended to cover the present invention any variations, uses, or adaptations, these modifications, purposes or
Person's adaptive change follows the general principle of the present invention and includes undocumented common knowledge in the art of the invention
Or conventional techniques.The description and examples are only to be considered as illustrative, and true scope and spirit of the invention are by appended
Claim is pointed out.
Claims (10)
1. a kind of malice occupy-place recognition methods of ticketing service order, which is characterized in that the method includes:
The user information and behavioural information that ticketing service order is completed in database are obtained, ticketing service order, which is completed, described in part is marked
It is denoted as malice occupy-place, extraction and the relevant characteristic variable of malice occupy-place from the user information and the behavioural information;
It in the value being completed in ticketing service order for being marked as malice occupy-place and is being not labeled as disliking based on each characteristic variable
The difference for the value of meaning occupy-place being completed in ticketing service order, calculates the value of information of each characteristic variable;
The characteristic variable that the value of information is more than information threshold is filtered out, malice occupy-place is being marked as based on the characteristic variable filtered out
The value being completed in ticketing service order, extract priori conditions;
Using characteristic variable training regression model Z=logit (p)=log (the odds)=β filtered out0+β1X1+β2X2+β3X3+…
+βkXk, wherein p is the malice occupy-place probability of a ticketing service order, and odds is that the ticketing service order occurs malice occupy-place and malice does not occur
The ratio of occupy-place, odds=p/ (1-p), X1,X2,…,XkIt is taken for each characteristic variable for filtering out is corresponding in the ticketing service order
Value, β0For intercept, β1,β2,…,βkFor the corresponding regression coefficient of each characteristic variable filtered out;
Real-time ticketing service order and its user information and behavioural information are obtained, judges user information and the behavior of the real-time ticketing service order
Whether information meets corresponding priori conditions, if then judging that the real-time ticketing service order has malice occupy-place risk;
For the real-time ticketing service order with malice occupy-place risk, the user information based on the real-time ticketing service order and behavior letter
Breath, the malice occupy-place probability of the real-time ticketing service order is calculated using trained regression model, and it is super to intercept malice occupy-place probability
Go out the real-time ticketing service order of probability threshold value.
2. the malice occupy-place recognition methods of ticketing service order as described in claim 1, which is characterized in that when training regression model,
Further include being adjusted to regression coefficient using following formula:
Wherein,For the lasso estimated values of regression coefficient;
N is the total sample number that ticketing service order is completed for training regression model, and i takes from 1 to N;
yiThe corresponding regression function of ticketing service order is completed for i-th,yi∈ { 0,1 }, yi=1
Represent ticketing service order malice occupy-place, yi=0 represents ticketing service order non-malicious occupy-place;
K is the characteristic variable number for training regression model filtered out, and j takes from 1 to k;
For penalty term;
T is binding occurrence.
3. the malice occupy-place recognition methods of ticketing service order as described in claim 1, which is characterized in that the interception malice occupy-place
Probability exceed probability threshold value real-time ticketing service order the step of include:
Malice occupy-place probability is divided into malice occupy-place grade according to multiple probability threshold values, to the real-time ticket of different malice occupy-place grades
The corresponding lower single user of business order, takes corresponding interception measure respectively.
4. the malice occupy-place recognition methods of ticketing service order as claimed in claim 3, which is characterized in that the interception measure packet
It includes:
It is less than the corresponding lower single user of real-time ticketing service order of the first probability threshold value for malice occupy-place probability, allows its ticketing service pre-
Determine behavior;
First probability threshold value is more than or equal to for malice occupy-place probability and is corresponded to less than the real-time ticketing service order of the second probability threshold value
Lower single user, pop up identifying code;
It is more than or equal to the corresponding lower single user of real-time ticketing service order of the second probability threshold value for malice occupy-place probability, refuses its ticket
Business reservation behavior.
5. the malice occupy-place recognition methods of ticketing service order as described in claim 1, which is characterized in that carry out priori conditions judgement
When, if a real-time ticketing service order is unsatisfactory for priori conditions, judges that the real-time ticketing service order does not have malice occupy-place risk, allow
Behavior is subscribed in the ticketing service of the corresponding lower single user of the real-time ticketing service order.
6. the malice occupy-place recognition methods of ticketing service order as described in claim 1, which is characterized in that obtain real-time ticketing service order
And its step of user information and behavioural information, includes:
Obtain real-time ticketing service order, and obtain the real-time ticketing service order correspond to described in the user information of characteristic variable that filters out and
Behavioural information.
7. the malice occupy-place recognition methods of ticketing service order as described in claim 1, which is characterized in that using the feature filtered out
Further include the steps that data prediction is carried out to the characteristic variable filtered out before variable trains regression model.
8. a kind of malice occupy-place identifying system of ticketing service order, which is characterized in that the system comprises:
Sample acquisition module obtains and the user information and behavioural information of ticketing service order is completed in database, complete described in part
It is marked as malice occupy-place at ticketing service order, is extracted from the user information and the behavioural information relevant with malice occupy-place
Characteristic variable;
Variable processing module, based on each characteristic variable the value being completed in ticketing service order for being marked as malice occupy-place and
It is not labeled as the difference for the value of malice occupy-place being completed in ticketing service order, calculates the value of information of each characteristic variable;
Priori processing module filters out the characteristic variable that the value of information is more than predetermined threshold value, based on the characteristic variable filtered out in quilt
Labeled as the value of malice occupy-place being completed in ticketing service order, priori conditions are extracted;
Model training module, using characteristic variable training regression model Z=logit (p)=log (the odds)=β filtered out0+β1X1+β2X2+β3X3+…+βkXk, wherein p is the malice occupy-place probability of a ticketing service order, and odds is that the ticketing service order occurs maliciously to account for
Position and the ratio that malice occupy-place does not occur, odds=p/ (1-p), X1,X2,…,XkIt is each characteristic variable for filtering out in the ticketing service
Corresponding value, β in order0For intercept, β1,β2,…,βkFor the corresponding regression coefficient of each characteristic variable filtered out;
Malice identification module obtains real-time ticketing service order and its user information and behavioural information, judges the real-time ticketing service order
Whether user information and behavioural information meet priori conditions, if then judging that the real-time ticketing service order has malice occupy-place risk;
Probability evaluation entity, for the real-time ticketing service order with malice occupy-place risk, the user based on the real-time ticketing service order
Information and behavioural information, the malice occupy-place probability of the real-time ticketing service order is calculated using trained regression model, and intercepts evil
Occupy-place probability of anticipating exceeds the real-time ticketing service order of probability threshold value.
9. a kind of malice occupy-place identification equipment of ticketing service order, which is characterized in that including:
Processor;And
Memory, the executable instruction for storing the processor;
Wherein, the processor is configured to carry out perform claim 1~7 any one of them of requirement via the execution executable instruction
The step of malice occupy-place identification of ticketing service order.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
The step of malice occupy-place identification of claim 1~7 any one of them ticketing service order is realized when execution.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711499025.0A CN108564423A (en) | 2017-12-28 | 2017-12-28 | Malice occupy-place recognition methods, system, equipment and the storage medium of ticketing service order |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711499025.0A CN108564423A (en) | 2017-12-28 | 2017-12-28 | Malice occupy-place recognition methods, system, equipment and the storage medium of ticketing service order |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108564423A true CN108564423A (en) | 2018-09-21 |
Family
ID=63530493
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711499025.0A Pending CN108564423A (en) | 2017-12-28 | 2017-12-28 | Malice occupy-place recognition methods, system, equipment and the storage medium of ticketing service order |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108564423A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109408556A (en) * | 2018-09-28 | 2019-03-01 | 中国平安人寿保险股份有限公司 | Abnormal user recognition methods and device, electronic equipment, medium based on big data |
CN110288350A (en) * | 2019-04-24 | 2019-09-27 | 武汉众邦银行股份有限公司 | User's Value Prediction Methods, device, equipment and storage medium |
CN111461815A (en) * | 2020-03-17 | 2020-07-28 | 上海携程国际旅行社有限公司 | Order recognition model generation method, recognition method, system, device and medium |
CN112163932A (en) * | 2020-09-30 | 2021-01-01 | 中国民航信息网络股份有限公司 | Malicious seat occupying order identification method and device and electronic equipment |
CN112396504A (en) * | 2021-01-21 | 2021-02-23 | 北京天通慧智科技有限公司 | E-commerce order intercepting method and device and electronic equipment |
CN112529509A (en) * | 2020-12-29 | 2021-03-19 | 安徽运通达物流科技有限公司 | Method for identifying malicious freight notes in transportation platform |
CN113328978A (en) * | 2020-02-28 | 2021-08-31 | 北京沃东天骏信息技术有限公司 | Malicious user identification method and device, computer storage medium and electronic equipment |
CN114881711A (en) * | 2022-07-11 | 2022-08-09 | 荣耀终端有限公司 | Method for carrying out anomaly analysis based on request behavior and electronic equipment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103458042A (en) * | 2013-09-10 | 2013-12-18 | 上海交通大学 | Microblog advertisement user detection method |
CN104809589A (en) * | 2015-05-08 | 2015-07-29 | 北京嘀嘀无限科技发展有限公司 | Order processing method and device |
CN106779126A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | Malice accounts for the processing method and system of an order |
CN106786560A (en) * | 2017-02-14 | 2017-05-31 | 中国电力科学研究院 | A kind of power system stability characteristic automatic extraction method and device |
CN107341716A (en) * | 2017-07-11 | 2017-11-10 | 北京奇艺世纪科技有限公司 | A kind of method, apparatus and electronic equipment of the identification of malice order |
-
2017
- 2017-12-28 CN CN201711499025.0A patent/CN108564423A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103458042A (en) * | 2013-09-10 | 2013-12-18 | 上海交通大学 | Microblog advertisement user detection method |
CN104809589A (en) * | 2015-05-08 | 2015-07-29 | 北京嘀嘀无限科技发展有限公司 | Order processing method and device |
CN106779126A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | Malice accounts for the processing method and system of an order |
CN106786560A (en) * | 2017-02-14 | 2017-05-31 | 中国电力科学研究院 | A kind of power system stability characteristic automatic extraction method and device |
CN107341716A (en) * | 2017-07-11 | 2017-11-10 | 北京奇艺世纪科技有限公司 | A kind of method, apparatus and electronic equipment of the identification of malice order |
Non-Patent Citations (2)
Title |
---|
丁颖等: "《学术影响力的测评 方法与实践》", 30 September 2017, 武汉大学出版社 * |
蔡楠: "《半参数平滑转换自回归模型理论研究及其应用》", 31 December 2016, 东北财经大学出版社 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109408556A (en) * | 2018-09-28 | 2019-03-01 | 中国平安人寿保险股份有限公司 | Abnormal user recognition methods and device, electronic equipment, medium based on big data |
CN109408556B (en) * | 2018-09-28 | 2024-02-02 | 中国平安人寿保险股份有限公司 | Abnormal user identification method and device based on big data, electronic equipment and medium |
CN110288350A (en) * | 2019-04-24 | 2019-09-27 | 武汉众邦银行股份有限公司 | User's Value Prediction Methods, device, equipment and storage medium |
CN113328978A (en) * | 2020-02-28 | 2021-08-31 | 北京沃东天骏信息技术有限公司 | Malicious user identification method and device, computer storage medium and electronic equipment |
CN113328978B (en) * | 2020-02-28 | 2023-06-27 | 北京沃东天骏信息技术有限公司 | Malicious user identification method and device, computer storage medium and electronic equipment |
CN111461815A (en) * | 2020-03-17 | 2020-07-28 | 上海携程国际旅行社有限公司 | Order recognition model generation method, recognition method, system, device and medium |
CN111461815B (en) * | 2020-03-17 | 2023-04-28 | 上海携程国际旅行社有限公司 | Order recognition model generation method, recognition method, system, equipment and medium |
CN112163932A (en) * | 2020-09-30 | 2021-01-01 | 中国民航信息网络股份有限公司 | Malicious seat occupying order identification method and device and electronic equipment |
CN112529509A (en) * | 2020-12-29 | 2021-03-19 | 安徽运通达物流科技有限公司 | Method for identifying malicious freight notes in transportation platform |
CN112396504A (en) * | 2021-01-21 | 2021-02-23 | 北京天通慧智科技有限公司 | E-commerce order intercepting method and device and electronic equipment |
CN114881711A (en) * | 2022-07-11 | 2022-08-09 | 荣耀终端有限公司 | Method for carrying out anomaly analysis based on request behavior and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108564423A (en) | Malice occupy-place recognition methods, system, equipment and the storage medium of ticketing service order | |
Usmani et al. | Stock market prediction using machine learning techniques | |
CN109241418A (en) | Abnormal user recognition methods and device, equipment, medium based on random forest | |
CN107766929B (en) | Model analysis method and device | |
US20190180379A1 (en) | Life insurance system with fully automated underwriting process for real-time underwriting and risk adjustment, and corresponding method thereof | |
US9122995B2 (en) | Classification of stream-based data using machine learning | |
CN112926699A (en) | Abnormal object identification method, device, equipment and storage medium | |
CN110163647A (en) | A kind of data processing method and device | |
CN110852881B (en) | Risk account identification method and device, electronic equipment and medium | |
CN109784779A (en) | Financial risk prediction technique, device and storage medium | |
CN107679734A (en) | It is a kind of to be used for the method and system without label data classification prediction | |
AU2019100968A4 (en) | A Credit Reporting Evaluation System Based on Mixed Machine Learning | |
CN108062674A (en) | Order fraud recognition methods, system, storage medium and electronic equipment based on GPS | |
CN112308462A (en) | Power consumer classification method and device | |
CN110147389A (en) | Account number treating method and apparatus, storage medium and electronic device | |
CN109784352A (en) | A kind of method and apparatus for assessing disaggregated model | |
CN111126629A (en) | Model generation method, system, device and medium for identifying brushing behavior | |
US11551104B2 (en) | Method and system for exchange of packets pertaining to an instrument | |
CN114187125A (en) | Claims case distribution method, device, equipment and storage medium | |
CN112634062B (en) | Hadoop-based data processing method, device, equipment and storage medium | |
CN117235608A (en) | Risk detection method, risk detection device, electronic equipment and storage medium | |
CN113420909A (en) | User response information prediction model establishing method and information prediction method | |
CN109117352B (en) | Server performance prediction method and device | |
CN109767333A (en) | Select based method, device, electronic equipment and computer readable storage medium | |
CN106779245B (en) | Event-based civil aviation demand prediction method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180921 |