CN106874741B - A kind of business same device identification method, calculates equipment and storage medium at device - Google Patents
A kind of business same device identification method, calculates equipment and storage medium at device Download PDFInfo
- Publication number
- CN106874741B CN106874741B CN201710225717.XA CN201710225717A CN106874741B CN 106874741 B CN106874741 B CN 106874741B CN 201710225717 A CN201710225717 A CN 201710225717A CN 106874741 B CN106874741 B CN 106874741B
- Authority
- CN
- China
- Prior art keywords
- cheating
- business information
- relationship
- sample data
- evaluation model
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/44—Program or device authentication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/50—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
- G06F21/57—Certifying or maintaining trusted computer platforms, e.g. secure boots or power-downs, version controls, system software checks, secure updates or assessing vulnerabilities
- G06F21/577—Assessing vulnerabilities and evaluating computer system security
Abstract
The present invention provides a kind of same device identification method of business, device, calculate equipment and storage medium, wherein, this method comprises: obtaining cheating sample data and non-cheating sample data, machine learning is carried out using cheating sample data and non-cheating sample data, establishes evaluation model；Two business information are obtained, according to evaluation model, determine whether two business information derive from same equipment.The present invention establishes evaluation model according to sample data, judges whether two business information derive from the same equipment online by the evaluation model, avoid relying on it is subsequent artificial it is counter look into, improve the accuracy and efficiency of identification.The evaluation model can be periodically updated simultaneously, more comprehensive to the defence of camouflage equipment, and malice one's share of expenses for a joint undertaking is difficult to get around the identification of evaluation model.
Description
Technical field
The present invention relates to Internet technical fields, in particular to a kind of same device identification method of business, device, meter
Calculate equipment and storage medium.
Background technique
Currently, malice one's share of expenses for a joint undertaking by modification equipment facility information or network environment information, by equipment disguise oneself as it is multiple not
Same equipment sends a large amount of business information and practises fraud to speculate.In order to safeguard the interests of normal users, need to identify not
Whether same business information derives from same equipment, to penetrate the camouflage of malice one's share of expenses for a joint undertaking.
Currently, generally according to the cheating case having confirmed that, manual service is counter look by way of identify that cheating business is believed
Breath.For the cheating case having confirmed that, technical staff analyzes the facility information of each equipment of malice one's share of expenses for a joint undertaking camouflage, network environment
Information and business information therefrom find out invariant all the same in the multiple equipment of camouflage, and the invariant group found out is combined into one
A new variable.Identify whether different business information comes from same equipment using the new variable.
The subsequent artificial anti-mode looked into of above-mentioned dependence determines a variable, identifies that cheating business is believed using determining variable
Breath, accuracy is very low, and the variable determined is difficult the pseudo- all devices taken on of covering, often can not for emerging invariant
Defence.And the case where easily causing multiple variable common defences, increase the time cost of management, leads to not effectively observe different changes
The performance situation of amount.
Summary of the invention
In view of this, the embodiment of the present invention be designed to provide a kind of same device identification method of business, device, calculating are set
Standby and storage medium judges whether two business information derive from the same equipment by the evaluation model online, avoids relying on
It is afterwards manually counter to look into, improve the accuracy and efficiency of identification.The evaluation model can be periodically updated simultaneously, be installed to puppet
Standby defence is more comprehensive, and malice one's share of expenses for a joint undertaking is difficult to get around the identification of evaluation model.
In a first aspect, the embodiment of the invention provides a kind of same device identification methods of business, which comprises
Cheating sample data and non-cheating sample data are obtained, the cheating sample data and the non-cheating sample are utilized
Data carry out machine learning, establish evaluation model；
Two business information are obtained, according to the evaluation model, it is same to determine whether described two business information derive from
Equipment.
With reference to first aspect, the embodiment of the invention provides the first possible implementation of above-mentioned first aspect,
In, it is described using the cheating sample data and the non-cheating sample data carries out machine learning, evaluation model is established, is wrapped
It includes:
Obtain the corresponding cheating relationship pair of the cheating sample data, and to obtain the non-cheating sample data corresponding non-
Cheating relationship pair；
Each cheating relationship is obtained to corresponding feature vector, and obtain each non-cheating relationship to corresponding feature to
Amount；
Using each cheating relationship to, each cheating relationship to corresponding feature vector, each non-work
Disadvantage relationship pair and each non-cheating relationship carry out machine learning to corresponding feature vector, obtain evaluation model.
The possible implementation of with reference to first aspect the first, the embodiment of the invention provides the of above-mentioned first aspect
Two kinds of possible implementations, wherein
It is described to obtain the corresponding cheating relationship pair of the cheating sample data, and obtain the non-cheating sample data and correspond to
Non- cheating relationship pair, comprising:
It is cheating relationship pair by the business information combination of two of same equipment is belonged in the cheating sample data；
It is non-cheating relationship pair by the business information combination of two of distinct device is belonged in the non-cheating sample data.
The possible implementation of with reference to first aspect the first, the embodiment of the invention provides the of above-mentioned first aspect
Three kinds of possible implementations, wherein it is described to obtain each cheating relationship to corresponding feature vector, and obtain each non-cheating
Relationship is to corresponding feature vector, comprising:
It is whether identical to the characteristic value of same feature in two business information for including to compare cheating relationship；If it is,
The first default value is set by the corresponding vector element value of the identical feature of characteristic value；If it is not, then characteristic value is different
The corresponding vector element value of feature is set as the second default value；By the cheating relationship to corresponding all vector element value structures
At the cheating relationship to corresponding feature vector；
It is whether identical to the characteristic value of same feature in two business information for including to compare non-cheating relationship；If so,
Then the first default value is set by the corresponding vector element value of the identical feature of characteristic value；If it is not, then by characteristic value difference
The corresponding vector element value of feature be set as the second default value；By the non-cheating relationship to corresponding all vector elements
Value constitutes the non-cheating relationship to corresponding feature vector.
With reference to first aspect, the embodiment of the invention provides the 4th kind of possible implementation of above-mentioned first aspect,
In, it is described according to the evaluation model, determine whether described two business information derive from same equipment, comprising:
According to the two of acquisition business information, the corresponding feature vector of described two business information is obtained；
According to described two business information and described eigenvector, determine that described two business are believed by the evaluation model
Whether breath derives from same equipment.
The 4th kind of possible implementation with reference to first aspect, the embodiment of the invention provides the of above-mentioned first aspect
Five kinds of possible implementations, wherein it is described according to described two business information and described eigenvector, pass through the evaluation mould
Type determines whether described two business information derive from same equipment, comprising:
According to described two business information and described eigenvector, by the evaluation model to described two business information
It gives a mark, obtains the corresponding homologous score of described two business information；
According to the homologous score and default score value, determine whether described two business information derive from same equipment.
The 4th kind of possible implementation with reference to first aspect, the embodiment of the invention provides the of above-mentioned first aspect
Six kinds of possible implementations, wherein it is corresponding to obtain described two business information for two business information according to acquisition
Feature vector, comprising:
Whether the characteristic value for comparing same feature in described two business information is identical；
If it is, in the corresponding feature vector of described two business information, the identical feature of characteristic value is corresponding
Vector element value is set as the first default value；
If it is not, then in the corresponding feature vector of described two business information, the different feature of characteristic value is corresponding
Vector element value is set as the second default value；
It is corresponding that the corresponding all vector element values of described two business information are constituted into described two business information
Feature vector.
The possible implementation of with reference to first aspect the first, the embodiment of the invention provides the of above-mentioned first aspect
Seven kinds of possible implementations, wherein it is described to obtain the corresponding cheating relationship pair of the cheating sample data, and obtain described non-
The corresponding non-cheating relationship pair of sample data of practising fraud, comprising:
Same equipment and the default consistent business information combination of two of feature will be belonged in the cheating sample data to make
Disadvantage relationship pair；
Distinct device will be belonged in the non-cheating sample data and the consistent business information of default feature group two-by-two
It is combined into non-cheating relationship pair.
The 7th kind of possible implementation with reference to first aspect, the embodiment of the invention provides the of above-mentioned first aspect
Eight kinds of possible implementations, wherein it is described to obtain each cheating relationship to corresponding feature vector, and obtain each non-cheating
Relationship is to corresponding feature vector, comprising:
Cheating relationship is compared to other features in two business information for including in addition to the default feature, described in acquisition
Cheating relationship is to corresponding feature vector；
Non- cheating relationship is compared to other features in two business information for including in addition to the default feature, obtains institute
Non- cheating relationship is stated to corresponding feature vector.
The 7th kind of possible implementation with reference to first aspect, the embodiment of the invention provides the of above-mentioned first aspect
Nine kinds of possible implementations, wherein
Two business information of the acquisition determine whether described two business information derive from according to the evaluation model
Same equipment, comprising:
Obtain consistent two business information of the default feature；
Other features in the described two business information obtained in addition to the default feature are compared, described two industry are obtained
The corresponding feature vector of information of being engaged in；
According to described two business information and described eigenvector, determine that described two business are believed by the evaluation model
Whether breath derives from same equipment.
With reference to first aspect and any implementation in first to the 9th kind of implementation of first aspect, the present invention
Embodiment provides the tenth kind of implementation of above-mentioned first aspect, wherein the method also includes:
Every preset time period obtains the cheating sample data in the past preset time period nearest from current time
With non-cheating sample data, according to the cheating sample data and the non-cheating sample data in the preset time period,
Update the evaluation model.
Second aspect, the embodiment of the invention provides a kind of same equipment identification device of business, described device includes:
Model building module utilizes the cheating sample number for obtaining cheating sample data and non-cheating sample data
Machine learning is carried out according to the non-cheating sample data, establishes evaluation model；
Module is obtained, for obtaining two business information；
Determining module, for determining whether described two business information derive from same equipment according to the evaluation model.
In conjunction with second aspect, the embodiment of the invention provides the first possible implementation of above-mentioned second aspect,
In, the model building module includes:
First acquisition unit, for obtaining the corresponding cheating relationship pair of the cheating sample data, and the acquisition non-work
The corresponding non-cheating relationship pair of disadvantage sample data；Each cheating relationship is obtained to corresponding feature vector, and obtains each non-work
Disadvantage relationship is to corresponding feature vector；
Machine learning unit, for using each cheating relationship to, each cheating relationship to corresponding feature
Vector, each non-cheating relationship pair and each non-cheating relationship carry out machine learning to corresponding feature vector, obtain
Obtain evaluation model.
In conjunction with second aspect, the embodiment of the invention provides second of possible implementation of above-mentioned second aspect,
In, the determining module includes:
Second acquisition unit obtains the corresponding spy of described two business information for two business information according to acquisition
Levy vector；
Determination unit, for being determined by the evaluation model according to described two business information and described eigenvector
Whether described two business information derive from same equipment.
The third aspect, the embodiment of the invention provides a kind of calculating equipment, including memory and processor, the memories
On be stored with the computer program that can be run on the processor, the processor executes when running the computer program
State method described in any implementation in first to the tenth kind of implementation of first aspect and first aspect.
Fourth aspect is stored with computer journey the embodiment of the invention provides a kind of storage medium on the storage medium
Sequence, the computer program execute first to the tenth kind of realization side of above-mentioned first aspect and first aspect when being run by processor
Method described in any implementation in formula.
In method and device provided in an embodiment of the present invention, cheating sample data and non-cheating sample data, benefit are obtained
Machine learning is carried out with cheating sample data and non-cheating sample data, establishes evaluation model；Two business information are obtained, according to
Evaluation model, determines whether two business information derive from same equipment.The present invention establishes evaluation model according to sample data,
Judge whether two business information derive from the same equipment online by the evaluation model, avoid relying on it is subsequent artificial it is counter look into,
Improve the accuracy and efficiency of identification.Simultaneously the evaluation model can periodically be updated, to camouflage equipment defence more
Comprehensively, malice one's share of expenses for a joint undertaking is difficult to get around the identification of evaluation model.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 shows a kind of business provided by the embodiment of the present invention 1 with the flow chart of device identification method；
Fig. 2 shows another kind business provided by the embodiment of the present invention 1 with the flow chart of device identification method；
Fig. 3 shows a kind of business provided by the embodiment of the present invention 2 with the first structure diagram of equipment identification device；
Fig. 4 shows a kind of business provided by the embodiment of the present invention 2 with the second structural schematic diagram of equipment identification device；
Fig. 5 shows a kind of structural schematic diagram for calculating equipment provided by the embodiment of the present invention 3.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
Middle attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
It is a part of the embodiment of the present invention, instead of all the embodiments.The present invention being usually described and illustrated herein in the accompanying drawings is real
The component for applying example can be arranged and be designed with a variety of different configurations.Therefore, of the invention to what is provided in the accompanying drawings below
The detailed description of embodiment is not intended to limit the range of claimed invention, but is merely representative of selected reality of the invention
Apply example.Based on the embodiment of the present invention, those skilled in the art institute obtained without making creative work
There are other embodiments, shall fall within the protection scope of the present invention.
A variable is determined in view of relying on the subsequent artificial anti-mode looked into the related technology, is known using determining variable
Not Zuo Bi business information, accuracy is very low, and the variable determined is difficult the pseudo- all devices taken on of covering, for it is emerging not
Variable can not often be defendd.And the case where easily causing multiple variable common defences, increase the time cost of management, leads to not have
Effect observes the performance situation of different variables.Based on this, the embodiment of the invention provides a kind of same device identification methods of business, dress
It sets, calculate equipment and storage medium, be described below by embodiment.
Embodiment 1
The embodiment of the invention provides a kind of same device identification methods of business.The resolution that malice one's share of expenses for a joint undertaking passes through modification equipment
One equipment is disguised oneself as multiple equipment, is practised fraud by the equipment pretended by the parameters such as rate, available machine time or remaining storage
Business is to speculate.The method provided through the embodiment of the present invention can judge automatically whether two business information derive from together
One equipment: it when the number of the business information from same equipment reaches certain value, determines from the same equipment
Business information is cheating business information, thus the equipment for penetrating the camouflage of malice one's share of expenses for a joint undertaking.
Referring to Fig. 1, this method specifically includes the following steps:
Step 101: obtaining cheating sample data and non-cheating sample data, utilize cheating sample data and non-cheating sample
Data carry out machine learning, establish evaluation model.
The cheating business information having confirmed that and non-cheating business information are obtained from past a large amount of business information, will acquire
All cheating business information composition cheating sample data, the non-cheating sample number of all non-cheating business information compositions that will acquire
According to.Wherein, cheating business information includes the multiple cheating business information for being had modified the same equipment of parameter by malice one's share of expenses for a joint undertaking and sending,
It also include multiple cheating business information that malice one's share of expenses for a joint undertaking is sent with the same equipment of unmodified parameter.Non- cheating business information is not
The non-cheating business information sent with equipment.
In embodiments of the present invention, cheating sample data and non-cheating sample data can be indicated with set.It is practising fraud
It include multiple subclass in the set of sample data, the element in each subclass is the business letter from same equipment
Breath.For example, the collection of cheating sample data is combined into { (S1,S2),(S3,S4,S5), business information S1And S2From same equipment M1,
Business information S3、S4And S5From another equipment M2.The collection of non-cheating sample data is combined into { S6,S7,S8, business information S6、S7
And S8It is respectively from different equipment, such as S6From equipment M3, S7From equipment M4, S8From equipment M5。
It is above-mentioned get cheating sample data and non-cheating sample data after, establish evaluation model by operating as follows:
The corresponding cheating relationship pair of cheating sample data is obtained, and obtains the corresponding non-cheating relationship of non-cheating sample data
It is right；Each cheating relationship is obtained to corresponding feature vector, and obtains each non-cheating relationship to corresponding feature vector；It utilizes
Each cheating relationship is to, each cheating relationship to corresponding feature vector, each non-cheating relationship pair and each non-cheating relationship
Machine learning is carried out to corresponding feature vector, obtains evaluation model.
It is cheating relationship pair by the business information combination of two of same equipment is belonged in sample data of practising fraud, by non-cheating sample
The business information combination of two for belonging to distinct device in notebook data is non-cheating relationship pair.
When obtaining cheating relationship to corresponding feature vector, cheating relationship is compared to same in two business information for including
Whether the characteristic value of one feature is identical；If it is, in the corresponding feature vector of two business information, characteristic value is identical
The corresponding vector element value of feature is set as the first default value；If it is not, then in the corresponding feature vector of two business information
In, the second default value is set by the corresponding vector element value of the different feature of characteristic value.Two business information are corresponding
All vector element values constitute the corresponding feature vector of the two business information.
In embodiments of the present invention, business information includes hardware information, the software information, network rings when equipment is traded
Border information and information relevant to transaction.Hardware information includes brand, model, resolution ratio and available machine time of equipment etc..Software
Information includes the information for the application program installed in equipment.Network environment information includes the http (hypertext of equipment
Transfer protocol, hypertext transfer protocol) address etc..Information relevant to transaction includes the shop that exchange is related to
Information, order information purchase by group odd numbers information etc..
When comparing cheating relationship to same feature in two business information for including, the feature to be compared includes upper
State the design parameter feature in hardware information, software information and network environment information, as resolution ratio, system time, the available machine time,
Residue storage and the address http etc..In embodiments of the present invention, the comparison sequence for presetting the feature successively compared, according to pre-
Whether the comparison sequence first set compares cheating relationship identical to the characteristic value of same feature in two business information for including, if
It is identical, the first default value is set by the corresponding vector element value of feature, is otherwise provided as the second default value.Compare in order
After complete all features, the corresponding vector element value of each feature is formed into the cheating relationship to corresponding feature vector.
Above-mentioned first default value can be 1 or 2 etc..Second default value is different from the first default value, can for 0 or
3 etc..The embodiment of the present invention does not limit the specific value of the first default value and the second default value, can root in practical application
The value of the first default value and the second default value is determined according to demand.
The operation of above-mentioned acquisition feature vector in order to facilitate understanding, citing is illustrated below.For example, it is assumed that cheating relationship
It include business information 1 and business information 2 to 1, the first default value is 1, and the second default value is 0.As shown in table 1, compare industry
Whether the characteristic value of same feature is identical in information 1 of being engaged in and business information 2, if it is, the corresponding vector element of setting feature
Value is the first default value 1, and it is the second default value 0 that the corresponding vector element of feature, which is otherwise arranged,.Each feature is corresponding
Vector element value forms cheating relationship to 1 corresponding feature vector (1,0,1,1)
Table 1
| Cheating relationship is to 1 | Feature 1 | Feature 2 | Feature 3 | Feature 4 |
| Business information 1- business information 2 | 1 | 0 | 1 | 1 |
Likewise it is possible to obtain each non-cheating in such a way that above-mentioned acquisition cheating relationship is to corresponding feature vector
Relationship is to corresponding feature vector.
Each cheating relationship is got through the above way to corresponding feature vector and each non-cheating relationship to correspondence
Feature vector after, using each cheating relationship to, each cheating relationship to corresponding feature vector, each non-cheating relationship pair
And each non-cheating relationship carries out machine learning to corresponding feature vector, establishes evaluation model by machine learning.By this
Evaluation model judges whether two business information derive from the same equipment, instead of the artificial subsequent anti-mode looked into, improves
The efficiency that business is identified with equipment.
In order to further increase the accuracy and coverage rate that business is identified with equipment, the embodiment of the present invention also passes through such as lower section
Formula regularly updates above-mentioned evaluation model, specifically includes:
Every preset time period obtains cheating sample data in the past preset time period nearest from current time and non-
Sample data of practising fraud updates evaluation model according to the cheating sample data and non-cheating sample data in preset time period.
In the embodiment of the present invention, according to the cheating sample data and non-cheating sample data in the preset time period of acquisition,
It is above-mentioned establish evaluation model by way of establish new evaluation model.Original evaluation model is replaced with into the new evaluation mould
Type.
Since machine learning mode every preset time period dynamic can be used to update evaluation model, so that the present invention is implemented
Example can timely respond to the fraudulent means of malice one's share of expenses for a joint undertaking, and malice one's share of expenses for a joint undertaking is difficult to get around the monitoring of evaluation model, improve cheating prison
The accuracy of control.
After establishing evaluation model by the operation of this step, on the server by evaluation model configuration, server is by being somebody's turn to do
Evaluation model, as follows 102 and 103 operation come monitor in real time the business information received whether derive from it is same
Equipment.
Step 102: obtain two business information, according to evaluation model, determine two business information obtaining whether source
In same equipment.
Two business information are obtained online.According to the two of acquisition business information, the corresponding spy of two business information is obtained
Levy vector；According to two business information and feature vector, it is same to determine whether two business information derive from by evaluation model
Equipment.
In the embodiment of the present invention, when receiving two business information online, same feature in the two business information is compared
Characteristic value it is whether identical；If it is, in the corresponding feature vector of two business information, by the identical feature pair of characteristic value
The vector element value answered is set as the first default value；If it is not, then in the corresponding feature vector of two business information, it will be special
The corresponding vector element value of the different feature of value indicative is set as the second default value.The corresponding institute of the two business information is oriented
Secondary element value constitutes the corresponding feature vector of the two business information.
After obtaining the corresponding feature vector of the two business information through the above way, according to the two business information and this
The corresponding feature vector of two business information gives a mark to the two business information by evaluation model, obtains the two industry
The corresponding homologous score of information of being engaged in；According to homologous score and default score value, determine whether two business information derive from same set
It is standby.
The two business information feature vector corresponding with the two business information is inputted into evaluation model, evaluation model is defeated
The corresponding homologous score of the two business information out, homologous score is for indicating the two business information from the same equipment
A possibility that, homologous score can be the numerical value more than or equal to 0 less than or equal to 1.Determine whether homologous score is greater than default score value,
If it is, determining that the two business information derive from same equipment, if it is not, then determining the two business information from not
Same equipment.
The embodiment of the present invention is determining two business information after same equipment, and also caching derives from the same equipment
Two business information, when receiving a new business information again, by the new business information and caching two industry
Any one business information in information of being engaged in forms business relations pair, obtains the business relations to corresponding feature vector, by the industry
Business relationship, in corresponding feature vector input evaluation model, obtains the industry to two business information and the business relations for including
The homologous score of business relationship pair, if the homologous score is greater than default score value, it is determined that two of the new business information and caching
Business information derives from the same equipment, the then store path of two business information before being cached to the new business information
Under.For each of subsequently received new business information all in accordance with aforesaid way determine new business information whether with before
Determining business information derives from the same equipment.
In the embodiment of the present invention, every preset time period, the interior business from same equipment of statistics preset time period is believed
The number of breath；When the number of statistics is greater than default value, determine the business information from same equipment for cheating business letter
Breath.
Due to malice one's share of expenses for a joint undertaking may within a very short time by the equipment pretended send a large amount of cheating business information to
Server, so if judging that the number of the business information derived from same equipment is greater than present count within a preset period of time
Value can then determine that the equipment is the equipment of malice one's share of expenses for a joint undertaking camouflage, and the business information which sends is cheating business information.
When determining that the business information from same equipment is cheating business information through the above way, not to these cheatings
Business information is responded, and the business information that the subsequent same equipment is sent also no longer is responded.
In embodiments of the present invention, evaluation model is obtained by machine learning, given a mark online to two business information,
Realize and same equipment Risk quantified, can with Robustness with the air control of equipment Risk, the accuracy rate of balance cheating identification and
Coverage rate.Using machine learning mode automatic identification, by modifying, hardware information, software is novel or the same of network environment information sets
Standby cheating business, the identification method for analyzing instead of artificial post-flight data, observing and manually setting improve recognition efficiency,
New cheat mode can be timely responded to, while anti-cheating system is from safeguarding that a plurality of rule becomes safeguarding an evaluation model, drop
Low maintenance cost.
Due to having some hardware informations and software information that can not usually modify, as malice one's share of expenses for a joint undertaking can not modify equipment
The information such as board and device model, so when obtaining the feature vector of two business information, the letter that these can not be modified
It ceases corresponding feature to foreclose, to reduce machine learning and the subsequent calculation amount judged online, shortens the time of machine learning,
And improve the efficiency of online recognition.
It is default feature by the corresponding feature-set of the above-mentioned information that can not be modified, in step 101 in the embodiment of the present invention
In establish evaluation model in the following way, specifically include:
Same equipment will be belonged in cheating sample data and the default consistent business information combination of two of feature is closed for cheating
System pair will belong to distinct device and the default consistent business information combination of two of feature for non-cheating pass in non-cheating sample data
System pair；Each cheating relationship is obtained to corresponding feature vector, and obtains each non-cheating relationship to corresponding feature vector；Benefit
, each cheating relationship closes corresponding feature vector, each non-cheating relationship pair and each non-cheating with each cheating relationship
System carries out machine learning to corresponding feature vector, obtains evaluation model.
When obtaining cheating relationship to corresponding feature vector, cheating relationship is compared to removing in two business information for including
Other features outside default feature obtain cheating relationship to corresponding feature vector.Non- cheating relationship is being obtained to corresponding spy
When levying vector, non-cheating relationship is compared to other features in two business information for including in addition to default feature, obtains non-work
Disadvantage relationship is to corresponding feature vector.
Above-mentioned default feature is the feature that equipment brand, device model etc. can not be modified, other in addition to default feature are special
Sign is the feature that resolution ratio, system time, available machine time, remaining storage and the address http etc. can be modified by malice one's share of expenses for a joint undertaking.
After establishing evaluation model through the above way, come to obtain two business online in a step 102 in the following way
Information and the corresponding feature vector of two business information of acquisition, comprising:
Two default consistent business information of feature are obtained online；It compares in two business information obtained except default feature
Other outer features obtain the corresponding feature vector of two business information.
Above-mentioned default consistent two business information of feature and the corresponding feature vector input of the two business information are commented
In valence model, the corresponding homologous score of the two business information is obtained.Judge whether the homologous score is greater than default score value, if
It is, it is determined that the two business information derive from the same equipment, if it is not, then determining the two business information from difference
Equipment.
In order to more intuitively understand method provided in an embodiment of the present invention, it is specifically described with reference to the accompanying drawing.Such as Fig. 2
It is shown, cheating relationship pair is constructed according to cheating sample data, non-cheating relationship pair is constructed according to non-cheating sample data, for making
Disadvantage relationship to non-cheating relationship pair, then it is trained to carry out machine learning algorithm to corresponding feature vector all acquisition relationships
To evaluation model.When online progress business is identified with equipment, two business information are obtained online, obtain the two business information pair
The feature vector answered, input evaluation model are given a mark, and evaluation model exports the corresponding homologous score of the two business information.
In embodiments of the present invention, the foundation of evaluation model can also be not based on cheating relationship to non-cheating relationship into
Row, but carry out machine learning to the cheating Transaction Information that has confirmed that establish evaluation model will receive when application on site
In the evaluation model that one business information input is established, to identify whether the business information is cheating business information.
The embodiment of the present invention can not also establish evaluation model, but according to cheating sample data, construct cheating relationship
It is right, then by a cheating feature vector at machine learning training, when online recognition, obtain two business information pair received
The feature vector answered, the similarity factor for calculating this feature vector between feature vector of practising fraud, when similarity factor reaches default phase
When like value, judge receive two business information from the same equipment.
In embodiments of the present invention, obtain cheating sample data and non-cheating sample data, using cheating sample data and
Non- cheating sample data carries out machine learning, establishes evaluation model；It obtains two business information and determines two according to evaluation model
Whether a business information derives from same equipment.The present invention establishes evaluation model according to sample data, passes through the evaluation model
Judge whether two business information derive from the same equipment online, avoid relying on it is subsequent artificial it is counter look into, improve the standard of identification
True property and efficiency.The evaluation model can be periodically updated simultaneously, more comprehensive to the defence of camouflage equipment, and malice one's share of expenses for a joint undertaking is very
Hardly possible gets around the identification of evaluation model.
Embodiment 2
Referring to Fig. 3, the embodiment of the invention provides a kind of same equipment identification device of business, the device is for executing above-mentioned reality
The same device identification method of business of the offer of example 1 is applied, which specifically includes:
Model building module 201 utilizes cheating sample data for obtaining cheating sample data and non-cheating sample data
Machine learning is carried out with non-cheating sample data, establishes evaluation model；
Module 202 is obtained, for obtaining two business information；
Determining module 203, for determining whether two business information derive from same equipment according to evaluation model.
As shown in figure 4, model building module 201 includes:
First acquisition unit 2011, for obtaining the corresponding cheating relationship pair of cheating sample data, and the non-cheating sample of acquisition
The corresponding non-cheating relationship pair of notebook data；Each cheating relationship is obtained to corresponding feature vector, and obtains each non-cheating and closes
System is to corresponding feature vector；
Machine learning unit 2012, for using each cheating relationship to, each cheating relationship to corresponding feature vector,
Each non-cheating relationship pair and each non-cheating relationship carry out machine learning to corresponding feature vector, obtain evaluation model.
Above-mentioned first acquisition unit 2011, for belonging to the business information of same equipment group two-by-two in the sample data that will practise fraud
It is combined into cheating relationship pair；It is non-cheating relationship by the business information combination of two of distinct device is belonged in non-cheating sample data
It is right.
As shown in figure 4, determining module 203 includes:
Second acquisition unit 2031 obtains the corresponding spy of two business information for two business information according to acquisition
Levy vector；
Determination unit 2032, for determining two business by evaluation model according to two business information and feature vector
Whether information derives from same equipment.
Above-mentioned determination unit 2032 is used for according to two business information and feature vector, by evaluation model to two industry
Business information is given a mark, and the corresponding homologous score of two business information is obtained；According to homologous score and default score value, two are determined
Whether business information derives from same equipment.
Whether above-mentioned second acquisition unit 2031, the characteristic value for comparing same feature in two business information are identical；
If it is, the corresponding vector element value of the identical feature of characteristic value is set in the corresponding feature vector of two business information
It is set to the first default value；If it is not, then in the corresponding feature vector of two business information, by the different feature pair of characteristic value
The vector element value answered is set as the second default value；The corresponding all vector element values of two business information are constituted into two industry
The corresponding feature vector of information of being engaged in.
Above-mentioned first acquisition unit 2011, for belonging to same equipment in the sample data that will practise fraud and default feature is consistent
Business information combination of two is cheating relationship pair；Distinct device and the default consistent industry of feature will be belonged in non-cheating sample data
Business information combination of two is non-cheating relationship pair.Compare cheating relationship in two business information for including in addition to default feature
Other features obtain cheating relationship to corresponding feature vector；Non- cheating relationship is compared to removing in two business information for including
Other features outside default feature, obtain non-cheating relationship to corresponding feature vector.
Determining module 203, for obtaining consistent two business information of default feature；Compare two business information obtained
In other features in addition to default feature, obtain the corresponding feature vector of two business information；According to two business information and spy
Vector is levied, determines whether two business information derive from same equipment by evaluation model.
In embodiments of the present invention, which further includes update module, obtains the past from working as every preset time period
Cheating sample data and non-cheating sample data in nearest preset time period of preceding time, according to the cheating in preset time period
Sample data and non-cheating sample data update evaluation model.
In embodiments of the present invention, obtain cheating sample data and non-cheating sample data, using cheating sample data and
Non- cheating sample data carries out machine learning, establishes evaluation model；It obtains two business information and determines two according to evaluation model
Whether a business information derives from same equipment.The present invention establishes evaluation model according to sample data, passes through the evaluation model
Judge whether two business information derive from the same equipment online, avoid relying on it is subsequent artificial it is counter look into, improve the standard of identification
True property and efficiency.The evaluation model can be periodically updated simultaneously, more comprehensive to the defence of camouflage equipment, and malice one's share of expenses for a joint undertaking is very
Hardly possible gets around the identification of evaluation model.
Embodiment 3
Referring to Fig. 5, the embodiment of the invention provides a kind of calculating equipment, which includes memory 301 and processing
Device 302, the computer program that can be run on processor 302 is stored on memory 301, and processor 302 runs the computer
When program, the same device identification method of business provided by above-described embodiment 1 is executed.
Specifically, memory 301 and processor 302 can be general memory and processor, not do specific limit here
It is fixed, it is connected between memory 301 and processor 302 by bus, when the computer of 302 run memory 301 of processor storage
When program, it is able to carry out the same device identification method of above-mentioned business, subsequent artificial counter looks into knowledge to solve to rely in the related technology
Not Zuo Bi business information, the variable for causing accuracy very low, and determining is difficult the pseudo- all devices taken on of covering, for newly occurring
Invariant the problem of can not often defending.It realizes and evaluation model is established according to sample data, sentenced online by the evaluation model
Whether disconnected two business information derive from the same equipment, avoid relying on it is subsequent it is artificial it is counter look into, improve identification accuracy and
Efficiency.The evaluation model can be periodically updated simultaneously, more comprehensive to the defence of camouflage equipment, and malice one's share of expenses for a joint undertaking is difficult to get around
The identification of evaluation model.
Embodiment 4
The embodiment of the invention provides a kind of storage medium, it is stored with computer program on the storage medium, the computer
When program is run by processor, the same device identification method of business that above-described embodiment 1 provides is executed.
Specifically, which can be general storage medium, such as mobile disk, hard disk, on the storage medium
Computer program when being run, the same device identification method of above-mentioned business is able to carry out, to solve to rely in the related technology
It is subsequent artificial it is counter look into identification cheating business information, the variable for causing accuracy very low, and determining be difficult covering it is pseudo- take on it is all
Equipment, the problem of can not often being defendd for emerging invariant.It realizes and evaluation model is established according to sample data, by this
Evaluation model judges whether two business information derive from the same equipment online, avoid relying on it is subsequent artificial it is counter look into, improve
The accuracy and efficiency of identification.The evaluation model can be periodically updated simultaneously, more comprehensive to the defence of camouflage equipment, be disliked
Meaning one's share of expenses for a joint undertaking is difficult to get around the identification of evaluation model.
Business provided by the embodiment of the present invention can be the specific hardware or installation in equipment with equipment identification device
In software or firmware etc. in equipment.The technical effect of device provided by the embodiment of the present invention, realization principle and generation and
Preceding method embodiment is identical, and to briefly describe, Installation practice part does not refer to place, can refer in preceding method embodiment
Corresponding contents.It is apparent to those skilled in the art that for convenience and simplicity of description, foregoing description is
The specific work process of system, device and unit, the corresponding process during reference can be made to the above method embodiment, it is no longer superfluous herein
It states.
In embodiment provided by the present invention, it should be understood that disclosed device and method, it can be by others side
Formula is realized.The apparatus embodiments described above are merely exemplary, for example, the division of the unit, only one kind are patrolled
Function division is collected, there may be another division manner in actual implementation, in another example, multiple units or components can combine or can
To be integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Coupling, direct-coupling or communication connection can be INDIRECT COUPLING or communication link by some communication interfaces, device or unit
It connects, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
In addition, each functional unit in embodiment provided by the invention can integrate in one processing unit, it can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing, in addition, term " the
One ", " second ", " third " etc. are only used for distinguishing description, are not understood to indicate or imply relative importance.
Finally, it should be noted that embodiment described above, only a specific embodiment of the invention, to illustrate the present invention
Technical solution, rather than its limitations, scope of protection of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair
It is bright to be described in detail, those skilled in the art should understand that: anyone skilled in the art
In the technical scope disclosed by the present invention, it can still modify to technical solution documented by previous embodiment or can be light
It is readily conceivable that variation or equivalent replacement of some of the technical features；And these modifications, variation or replacement, do not make
The essence of corresponding technical solution is detached from the spirit and scope of technical solution of the embodiment of the present invention.Should all it cover in protection of the invention
Within the scope of.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (14)
1. a kind of same device identification method of business, which is characterized in that the described method includes:
Cheating sample data and non-cheating sample data are obtained, the cheating sample data and the non-cheating sample data are utilized
Machine learning is carried out, evaluation model is established；
Two business information are obtained, according to the evaluation model, determine whether described two business information derive from same equipment；
It is described using the cheating sample data and the non-cheating sample data carries out machine learning, establish evaluation model, wrap
It includes:
The corresponding cheating relationship pair of the cheating sample data is obtained, and obtains the corresponding non-cheating of the non-cheating sample data
Relationship pair；
Each cheating relationship is obtained to corresponding feature vector, and obtains each non-cheating relationship to corresponding feature vector；
, each cheating relationship closes corresponding feature vector, each non-cheating using each cheating relationship
System pair and each non-cheating relationship carry out machine learning to corresponding feature vector, obtain evaluation model.
2. the method according to claim 1, wherein described obtain the corresponding cheating pass of the cheating sample data
System pair, and obtain the corresponding non-cheating relationship pair of the non-cheating sample data, comprising:
It is cheating relationship pair by the business information combination of two of same equipment is belonged in the cheating sample data；
It is non-cheating relationship pair by the business information combination of two of distinct device is belonged in the non-cheating sample data.
3. the method according to claim 1, wherein it is described obtain each cheating relationship to corresponding feature to
Amount, and each non-cheating relationship is obtained to corresponding feature vector, comprising:
It is whether identical to the characteristic value of same feature in two business information for including to compare cheating relationship；If it is, by special
The corresponding vector element value of the identical feature of value indicative is set as the first default value；If it is not, then the feature that characteristic value is different
Corresponding vector element value is set as the second default value；The cheating relationship is constituted into institute to corresponding all vector element values
Cheating relationship is stated to corresponding feature vector；
It is whether identical to the characteristic value of same feature in two business information for including to compare non-cheating relationship；If it is, will
The corresponding vector element value of the identical feature of characteristic value is set as the first default value；If it is not, then the spy that characteristic value is different
It levies corresponding vector element value and is set as the second default value；By the non-cheating relationship to corresponding all vector element value structures
At the non-cheating relationship to corresponding feature vector.
4. determining described two industry the method according to claim 1, wherein described according to the evaluation model
Whether business information derives from same equipment, comprising:
According to the two of acquisition business information, the corresponding feature vector of described two business information is obtained；
According to described two business information and described eigenvector, determine that described two business information are by the evaluation model
It is no to derive from same equipment.
5. according to the method described in claim 4, it is characterized in that, it is described according to described two business information and the feature to
Amount, determines whether described two business information derive from same equipment by the evaluation model, comprising:
According to described two business information and described eigenvector, described two business information are carried out by the evaluation model
Marking, obtains the corresponding homologous score of described two business information；
According to the homologous score and default score value, determine whether described two business information derive from same equipment.
6. according to the method described in claim 4, it is characterized in that, two business information according to acquisition, described in acquisition
The corresponding feature vector of two business information, comprising:
Whether the characteristic value for comparing same feature in described two business information is identical；
If it is, in the corresponding feature vector of described two business information, by the corresponding vector of the identical feature of characteristic value
Element value is set as the first default value；
If it is not, then in the corresponding feature vector of described two business information, by the corresponding vector of the different feature of characteristic value
Element value is set as the second default value；
The corresponding all vector element values of described two business information are constituted into the corresponding feature of described two business information
Vector.
7. the method according to claim 1, wherein described obtain the corresponding cheating pass of the cheating sample data
System pair, and obtain the corresponding non-cheating relationship pair of the non-cheating sample data, comprising:
Same equipment will be belonged in the cheating sample data and the default consistent business information combination of two of feature is closed for cheating
System pair；It wherein, is default feature by the corresponding feature-set of information that can not be modified in the hardware information of equipment and software information；
Distinct device will be belonged in the non-cheating sample data and the consistent business information combination of two of the default feature is
Non- cheating relationship pair.
8. the method according to the description of claim 7 is characterized in that it is described obtain each cheating relationship to corresponding feature to
Amount, and each non-cheating relationship is obtained to corresponding feature vector, comprising:
Cheating relationship is compared to other features in two business information for including in addition to the default feature, obtains the cheating
Relationship is to corresponding feature vector；
Non- cheating relationship is compared to other features in two business information for including in addition to the default feature, is obtained described non-
Cheating relationship is to corresponding feature vector.
9. the method according to the description of claim 7 is characterized in that two business information of the acquisition, according to the evaluation mould
Type, determines whether described two business information derive from same equipment, comprising:
Obtain consistent two business information of the default feature；
Other features in the described two business information obtained in addition to the default feature are compared, described two business letters are obtained
Cease corresponding feature vector；
According to described two business information and described eigenvector, determine that described two business information are by the evaluation model
It is no to derive from same equipment.
10. -9 described in any item methods according to claim 1, which is characterized in that the method also includes:
Every preset time period obtains cheating sample data in the past preset time period nearest from current time and non-
Sample data of practising fraud is updated according to the cheating sample data and the non-cheating sample data in the preset time period
The evaluation model.
11. a kind of same equipment identification device of business, which is characterized in that described device includes:
Model building module, for obtaining cheating sample data and non-cheating sample data, using the cheating sample data with
The non-cheating sample data carries out machine learning, establishes evaluation model；
Module is obtained, for obtaining two business information；
Determining module, for determining whether described two business information derive from same equipment according to the evaluation model；
The model building module includes:
First acquisition unit, for obtaining the corresponding cheating relationship pair of the cheating sample data, and the acquisition non-cheating sample
The corresponding non-cheating relationship pair of notebook data；Each cheating relationship is obtained to corresponding feature vector, and obtains each non-cheating and closes
System is to corresponding feature vector；
Machine learning unit, for using each cheating relationship to, each cheating relationship to corresponding feature vector,
Each non-cheating relationship pair and each non-cheating relationship carry out machine learning to corresponding feature vector, are evaluated
Model.
12. device according to claim 11, which is characterized in that the determining module includes:
Second acquisition unit, for two business information according to acquisition, obtain the corresponding feature of described two business information to
Amount；
Determination unit is used for according to described two business information and described eigenvector, by described in evaluation model determination
Whether two business information derive from same equipment.
13. a kind of calculating equipment, including memory and processor, it is stored with and can runs on the processor on the memory
Computer program, which is characterized in that the processor executes the claims 1 to 10 when running the computer program
Method described in one.
14. a kind of storage medium, computer program is stored on the storage medium, which is characterized in that the computer program
The described in any item methods of the claims 1 to 10 are executed when being run by processor.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2017100041749 | 2017-01-04 | ||
| CN201710004174 | 2017-01-04 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106874741A CN106874741A (en) | 2017-06-20 |
| CN106874741B true CN106874741B (en) | 2019-03-29 |
Family
ID=59160863
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710225717.XA Active CN106874741B (en) | 2017-01-04 | 2017-04-07 | A kind of business same device identification method, calculates equipment and storage medium at device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106874741B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110928537B (en) * | 2018-09-19 | 2023-08-11 | 百度在线网络技术（北京）有限公司 | Model evaluation method, device, equipment and computer readable medium |
| CN110543756B (en) * | 2019-09-05 | 2021-05-11 | 同盾控股有限公司 | Device identification method and device, storage medium and electronic device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102163251A (en) * | 2010-02-22 | 2011-08-24 | 深圳市腾讯计算机系统有限公司 | Method and device for recognizing game cheating |
| CN105279674A (en) * | 2015-10-13 | 2016-01-27 | 精硕世纪科技（北京）有限公司 | Method and device for determining cheating behaviors of mobile advertisement delivering device |
| CN105677869A (en) * | 2016-01-06 | 2016-06-15 | 广州神马移动信息科技有限公司 | Multidimensional search log anti-cheating method, system and computing equipment |
| CN105812195A (en) * | 2014-12-30 | 2016-07-27 | 阿里巴巴集团控股有限公司 | Method and device for computer to identify batch accounts |
| CN106096974A (en) * | 2016-06-02 | 2016-11-09 | 中国联合网络通信集团有限公司 | A kind of anti-cheat method for shopping at network and system |
-
2017
- 2017-04-07 CN CN201710225717.XA patent/CN106874741B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102163251A (en) * | 2010-02-22 | 2011-08-24 | 深圳市腾讯计算机系统有限公司 | Method and device for recognizing game cheating |
| CN105812195A (en) * | 2014-12-30 | 2016-07-27 | 阿里巴巴集团控股有限公司 | Method and device for computer to identify batch accounts |
| CN105279674A (en) * | 2015-10-13 | 2016-01-27 | 精硕世纪科技（北京）有限公司 | Method and device for determining cheating behaviors of mobile advertisement delivering device |
| CN105677869A (en) * | 2016-01-06 | 2016-06-15 | 广州神马移动信息科技有限公司 | Multidimensional search log anti-cheating method, system and computing equipment |
| CN106096974A (en) * | 2016-06-02 | 2016-11-09 | 中国联合网络通信集团有限公司 | A kind of anti-cheat method for shopping at network and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106874741A (en) | 2017-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108564460B (en) | Real-time fraud detection method and device in internet credit scene | |
| CN107194623A (en) | A kind of discovery method and device of clique's fraud | |
| CN109583904A (en) | Training method, impaired operation detection method and the device of abnormal operation detection model | |
| CN109464807A (en) | Detect game plug-in method, apparatus and terminal | |
| CN107480660A (en) | Dangerous goods identifying system and its method | |
| CN104866831B (en) | The face recognition algorithms of characteristic weighing | |
| CN107404656A (en) | Live video recommends method, apparatus and server | |
| CN110798709B (en) | Video processing method and device, storage medium and electronic device | |
| CN105224775A (en) | Based on the method and apparatus that picture processing is arranged in pairs or groups to clothes | |
| CN108429718A (en) | Account recognition methods and device | |
| CN106874741B (en) | A kind of business same device identification method, calculates equipment and storage medium at device | |
| CN107038604A (en) | The methods of exhibiting and device of product object number of users | |
| CN109309809A (en) | The method and data processing method, device and system of trans-regional target trajectory tracking | |
| CN108985954A (en) | A kind of method and relevant device of incidence relation that establishing each mark | |
| CN108921209A (en) | Image identification method, device and electronic equipment | |
| CN109840778A (en) | The recognition methods of fraudulent user and device, readable storage medium storing program for executing | |
| CN108319974A (en) | Data processing method, device, storage medium and electronic device | |
| KR20190106347A (en) | SYSTEM FOR PROVIDING CUSTOMIZED WELCOME DATA BY RECOGNIZING CUSTOMER USING IoT DEVICES | |
| CN112200263B (en) | Self-organizing federal clustering method applied to power distribution internet of things | |
| CN111444075B (en) | Method for automatically discovering key influence indexes | |
| GB2604787A (en) | Systems and methods for objective-based skill training | |
| CN110503022A (en) | A kind of personal identification method, apparatus and system | |
| CN106529456A (en) | Information matching and information transmitting/receiving method, device and target object finding system | |
| CN109299884A (en) | A kind of influence power appraisal procedure and assessment device | |
| CN110008980A (en) | Identification model generation method, recognition methods, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |