CN109285009A - It brushes single recognition methods and brushes single identification device - Google Patents
It brushes single recognition methods and brushes single identification device Download PDFInfo
- Publication number
- CN109285009A CN109285009A CN201810886770.9A CN201810886770A CN109285009A CN 109285009 A CN109285009 A CN 109285009A CN 201810886770 A CN201810886770 A CN 201810886770A CN 109285009 A CN109285009 A CN 109285009A
- Authority
- CN
- China
- Prior art keywords
- subset
- cluster
- user
- brush
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
Landscapes
- Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (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 application provides a kind of single recognition methods of brush, comprising: according to the order information of first kind user in the first set of the single first kind user of doubtful brush, determines the second set of the Second Type user of doubtful cooperation brush list；According to the order information of Second Type user in the second set, the first kind user in the first set is extended；First set and second set are clustered respectively, obtain the first subset and second subset；Determine the first kind user and Second Type user that cooperation brush is single in first subset and the second subset.In accordance with an embodiment of the present disclosure, the set that can be constituted to user identifies, is beneficial to improve recognition efficiency, and be also able to carry out and efficiently identify for complicated group's cooperation brush single game scape.
Description
Technical field
This application involves technical field of data processing, in particular to the single recognition methods of brush, brush single identification device, electricity
Sub- equipment and computer readable storage medium.
Background technique
The mode of identification net about vehicle brush list at present, mainly identifies according to order feature and driver's portrait, this judgement
Mode is judged that one side efficiency is lower primarily directed to individual goal, on the other hand single for complicated group's cooperation brush
Scene is difficult to efficiently identify.
Summary of the invention
A kind of single recognition methods of brush is proposed according to a first aspect of the embodiments of the present invention, comprising:
According to the order information of first kind user in the first set of the single first kind user of doubtful brush, determine doubtful
The second set of the single Second Type user of cooperation brush；
According to the order information of Second Type user in the second set, the first kind in the first set is extended
User；
First set and second set are clustered respectively, obtain the first subset and second subset；
Determine the first kind user and Second Type user that cooperation brush is single in first subset and the second subset.
Optionally, the method also includes:
According to the order information of the first kind user in the first set of i-th extension, i-th extension second collection
Second Type user in conjunction, i are positive integer；
According to the order information of the Second Type user in the second set of i-th extension, i+1 time extension described first
First kind user in set；
First set to i+1 time extension and the Second Type user in the second set of i extension cluster, and obtain
To the first subset of i+1 time cluster and the second subset of i-th cluster；
Whether the second subset of the first subset and i-th cluster that determine i+1 time cluster restrains；
If the second subset convergence of the first subset and i-th cluster of i+1 time cluster, the of i+1 time cluster is determined
Cooperation brush single first kind user and Second Type user in the second subset of one subset and i-th cluster；
Wherein, however, it is determined that cooperate brush single first kind user and Second Type user out, the first son that i+1 times is clustered
Collect the first set extended as i-th, i is increased 1, according to the first kind user's in the first set of i-th extension
Order information, i-th extend the Second Type user in the second set, until the first subset or i-th of i+1 time cluster
The second subset of secondary cluster restrains, or until can not determine cooperation brush single first kind user and Second Type user.
Optionally, whether the second subset of the first subset of the determining i+1 time cluster and i-th cluster restrains packet
It includes:
Whether the first subset of the first subset and i-th cluster that determine i+1 time cluster is identical and i-th clusters
Second subset and (i-1)-th time cluster second subset it is whether identical；
If the first subset that the first subset of i+1 time cluster and i-th cluster is identical, and the second son of i-th cluster
Collect identical with the second subset that (i-1)-th time clusters, determines the first subset of i+1 time cluster and the second subset of i-th cluster
Convergence.
Optionally, the Second Type user in the first set of i+1 time extension and the second set of i extension
It is clustered, the second subset of the first subset and i-th cluster that obtain i+1 time cluster includes:
The first kind user in the first set of i+1 time extension is carried out according to the characteristic information of first kind user
Spatial characterization, the Second Type user in the second set extended according to the characteristic information of Second Type user to i-th carry out
Spatial characterization；
First between every two first kind user is being determined in the first set of i+1 time extension by clustering algorithm
Distance, and the second distance between every two Second Type user is being determined in the second set of i-th extension；
The first subset of i+1 time cluster is constituted less than the first kind user of the first pre-determined distance based on first distance,
The second subset of i-th cluster is constituted less than the Second Type user of the second pre-determined distance based on second distance.
Optionally, cooperate brush list in the first subset of the secondary cluster of the determining i+1 and the second subset of i-th cluster
First kind user and Second Type user include:
Based on the verifying model obtained by machine learning, the user and i-th in the first subset of i+1 time cluster is verified
It is single whether the user in the second subset of secondary cluster cooperates brush.
According to a second aspect of the embodiments of the present invention, a kind of single identification device of brush is proposed, comprising:
Gather determining module, for first kind user in the first set according to the single first kind user of doubtful brush
Order information determines the second set of the Second Type user of doubtful cooperation brush list；
Set expansion module extends described for the order information according to Second Type user in the second set
First kind user in one set；
Cluster module obtains the first subset and second subset for clustering respectively to first set and second set；
Single determining module is brushed, the first kind for determining that cooperation brush is single in first subset and the second subset is used
Family and Second Type user.
Optionally, the first kind user's in first set that the set expansion module is used to be extended according to i-th
Order information, i-th extend the Second Type user in the second set, and i is positive integer；And according to i-th extension
The order information of Second Type user in second set, i+1 time extend the first kind user in the first set；
The cluster module is used for the second class in the first set of i+1 time extension and the second set of i extension
Type user clusters, and obtains the first subset of i+1 time cluster and the second subset of i-th cluster；
Described device further include:
Determining module is restrained, whether the second subset that the first subset and i-th for determining i+1 time cluster cluster is received
It holds back；
Wherein, the second subset convergence brushing single determining module and being clustered in the first subset and i-th of i+1 time cluster
In the case where, the first kind for determining that cooperation brush is single in the first subset of i+1 time cluster and the second subset of i-th cluster is used
Family and Second Type user；
Wherein, however, it is determined that cooperate brush single first kind user and Second Type user out, the first son that i+1 times is clustered
Collect the first set extended as i-th, i is increased 1, the first collection that the set expansion module is used to extend according to i-th
The order information of first kind user in conjunction, i-th extend the Second Type user in the second set, until i+1
The second subset convergence of the first subset or the i-th cluster of secondary cluster, or until can not determine the single first kind of cooperation brush
User and Second Type user.
Optionally, the convergence determining module includes:
Identical determining module, for determine i+1 time cluster the first subset and i-th cluster the first subset whether phase
Together and i-th cluster second subset and (i-1)-th time cluster second subset it is whether identical；
Wherein, if the first subset for cluster of the first subset and i-th of i+1 time cluster is identical, and i-th cluster the
The second subset that two subsets are clustered with (i-1)-th time is identical, determines the first subset of i+1 time cluster and i-th cluster second
Subset convergence.
Optionally, the cluster module includes:
Spatial characterization submodule, the first set that i+1 time is extended for the characteristic information according to first kind user
In first kind user carry out spatial characterization, the second set that i-th is extended according to the characteristic information of Second Type user
In Second Type user carry out spatial characterization；
Distance determines submodule, for determining every two the in the first set of i+1 time extension by clustering algorithm
First distance between one type of user, and i-th extension second set in determine every two Second Type user it
Between second distance；
User cluster submodule, for based on first distance less than the first pre-determined distance first kind user constitute i-th+
First subset of 1 cluster constitutes the of i-th cluster less than the Second Type user of the second pre-determined distance based on second distance
Two subsets.
Optionally, the single determining module of the brush includes:
Submodule is verified, for verifying the first son of i+1 time cluster based on the verifying model obtained by machine learning
It is single whether the user in the second subset of the user of concentration and i-th cluster cooperates brush.
According to a third aspect of the embodiments of the present invention, a kind of electronic equipment is proposed, comprising:
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to executing the step in the single recognition methods of brush described in any of the above-described embodiment.
According to a fourth aspect of the embodiments of the present invention, it proposes a kind of computer readable storage medium, is stored thereon with calculating
Machine program, the program execute the step in the single recognition methods of brush described in any of the above-described embodiment when being executed by processor.
According to an embodiment of the invention, relative to the case where individual is identified is directed in the related technology, it can be to user
The set of composition identified, is beneficial to improve recognition efficiency, and also can be into for complicated group's cooperation brush single game scape
Row efficiently identifies.
Detailed description of the invention
Fig. 1 is a kind of schematic flow diagram of brush list recognition methods shown in embodiment according to the present invention.
Fig. 2 is another schematic flow diagram for brushing single recognition methods shown in embodiment according to the present invention.
Fig. 3 is the schematic flow diagram of the single recognition methods of another brush shown in embodiment according to the present invention.
Fig. 4 is the schematic flow diagram of the single recognition methods of another brush shown in embodiment according to the present invention.
Fig. 5 is the schematic flow diagram of the single recognition methods of another brush shown in embodiment according to the present invention.
Fig. 6 is a kind of hardware structure diagram of the single identification device place server of brush shown in embodiment according to the present invention.
Fig. 7 is a kind of schematic block diagram of brush list identification device shown in embodiment according to the present invention.
Fig. 8 is another schematic block diagram for brushing single identification device shown in embodiment according to the present invention.
Fig. 9 is a kind of schematic block diagram of convergence determining module shown in embodiment according to the present invention.
Figure 10 is a kind of schematic block diagram of cluster module shown in embodiment according to the present invention.
Figure 11 is a kind of schematic block diagram of brush list determining module shown in embodiment according to the present invention.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the application.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects be described in detail in claims, the application.
It is only to be not intended to be limiting the application merely for for the purpose of describing particular embodiments in term used in this application.
It is also intended in the application and the "an" of singular used in the attached claims, " described " and "the" including majority
Form, unless the context clearly indicates other meaning.It is also understood that term "and/or" used herein refers to and wraps
It may be combined containing one or more associated any or all of project listed.
It will be appreciated that though various information, but this may be described using term first, second, third, etc. in the application
A little information should not necessarily be limited by these terms.These terms are only used to for same type of information being distinguished from each other out.For example, not departing from
In the case where the application range, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as
One information.Depending on context, word as used in this " if " can be construed to " ... when " or " when ...
When " or " in response to determination ".
Fig. 1 is a kind of schematic flow diagram of brush list recognition methods shown in embodiment according to the present invention.The present embodiment institute
The single recognition methods of the brush shown can be applied to the scene that at least two class users trade.Such as net about parking lot scape, one type
User is driver, and another kind of user is passenger；Such as shopping at network scene, one type user are businessmans, another kind of user is
Customer.Certainly, the applicable scene of method shown in the present embodiment is not limited to above-mentioned scene, specifically can according to need progress
Setting.Mainly the embodiment of the present invention is illustrated under net about parking lot scape below.
As shown in Figure 1, the single recognition methods of the brush may comprise steps of:
Step S1, according to the order information of first kind user in the first set of the single first kind user of doubtful brush,
Determine the second set of the Second Type user of doubtful cooperation brush list.
In one embodiment, if first kind user is driver, then Second Type user is passenger；If the first kind
Type user is passenger, then Second Type user is then driver.It in first kind user is mainly below driver, Second Type is used
In the case that family is passenger, the embodiment of the present invention is illustrated.
It in one embodiment, can be according to the routing information of the completion order of driver, order frequency for driver
Information, order subsidize information etc. and determine the single driver of doubtful brush.
Such as by taking routing information as an example, if the order for completing certain paths accounts for blanket order in the order that driver completes
Large percentage, then can determine that driver is the single driver of doubtful brush.
Such as by taking order frequency information as an example, if the quantity on order completed in driver's unit time is more, then can be true
Determining driver is the single driver of doubtful brush.
Such as by taking order subsidizes information as an example, if the order for obtaining subsidy accounts for blanket order in the order that driver completes
Large percentage, then can determine that driver is the single driver of doubtful brush.
It should be noted that it is above-mentioned several to determine that information based on the single first kind user of doubtful brush is not limited to,
It can according to need and be configured.It and can be every when determining the single first kind user of doubtful brush according to much information
Weight is respectively set in kind information.
In one embodiment, the driver of doubtful brush list determined can be one, be also possible to multiple, and what is determined is doubtful
The single driver of brush may be constructed first set.It, can be according to the order information of driver, really for each driver in first set
The passenger of fixed doubtful cooperation brush list, the passenger of the doubtful cooperation brush list determined may be constructed second set.Brush doubtful for one
For single driver, the passenger of the doubtful cooperation brush list determined can be one, or multiple.It is mainly single in doubtful brush below
Driver be one in the case where, the embodiment of the present invention is illustrated.
In one embodiment, according to the order information of driver, the single passenger of cooperation brush is determined, wherein the order of driver
Information may include at least one of: portrait information, routing information, order frequency information, and order subsidizes information, evaluation letter
Breath.
Such as by taking routing information as an example, if completing the quantity of the order of certain paths in the order that driver is completed, with
In the order that passenger is completed, the quantity for completing the order of certain paths is close, then can determine that the passenger is single for cooperation brush
Passenger.
Such as by taking order frequency information as an example, if being completed in some unit time period in the order that driver is completed
The quantity of order, in the order completed with passenger, the quantity that order is completed in some unit time period is close, then can be with
Determine that the passenger brushes single passenger for cooperation.
It should be noted that it is above-mentioned several to determine that information based on the single Second Type user of cooperation brush is not limited to,
It can according to need and be configured.It and can be every when determining the single Second Type user of cooperation brush according to much information
Weight is respectively set in kind information.
Step S2 extends in the first set according to the order information of Second Type user in the second set
First kind user.
In one embodiment, after the second set of passenger of doubtful cooperation brush list has been determined, in second set
The passenger of each doubtful cooperation brush list can further determine the driver of doubtful cooperation brush list according to passenger's order information, for
The driver of doubtful cooperation brush list, can expand first set determined by each second passenger, increase in first set
The driver of doubtful cooperation brush list.
Wherein, the order information of passenger may include at least one of: portrait information, routing information, order frequency letter
Breath, order subsidize information, evaluation information.
Such as by taking routing information as an example, if completing the quantity of the order of certain paths in the order that passenger is completed, with
In the order that driver is completed, the quantity for completing the order of certain paths is close, then can determine that the driver is doubtful cooperation
Brush single driver.
Such as by taking order frequency information as an example, if being completed in some unit time period in the order that passenger is completed
The quantity of order, in the order completed with driver, the quantity that order is completed in some unit time period is close, then can be with
Determine that the driver is doubtful cooperation brush single driver.
Step S3, respectively clusters first set and second set, obtains the first subset and second subset.
It due to the driver single for doubtful brush of driver in first set, namely may both include cooperating brush list with passenger really
Driver also includes cooperating the single driver of brush not with passenger, and similarly, second set is also may be both including cooperating to brush with driver really
Single passenger also includes cooperating the single passenger of brush not with driver.Therefore, if directly determining conjunction based on first set and second set
Make brush single passenger and driver, then by larger processing may not be only resulted in view of not cooperating the single driver and passenger of brush
Data volume increases, it is also possible to which it is single that the driver and passenger that will do not cooperate brush list is mistaken for cooperation brush.
In one embodiment, spatial characterization can be carried out to the driver in first set according to the characteristic information of driver,
And spatial characterization is carried out to the passenger in second set according to the characteristic information of passenger, then clustering to first set can
To obtain the first subset, second set is carried out to cluster available second subset, the element in subset after clustering, relatively
The element in set before cluster is closer between element namely characteristic information is closer to.
And characteristic information is the characteristic information with cooperation brush simple correlation, more close characteristic information illustrates in the first subset
Driver brush single characteristic and become apparent, illustrate the brush list characteristic of driver in the first subset become apparent namely the first subset in
Driver cooperates the single mathematical expectation of probability of brush, and it is bigger to cooperate the single mathematical expectation of probability of brush relative to the driver in first set.Similarly, second
Passenger in subset cooperates the single mathematical expectation of probability of brush, and it is bigger to cooperate the single mathematical expectation of probability of brush relative to the passenger in second set.
Accordingly, subsequent that the single passenger and driver of cooperation brush can be determined based on the first subset and second subset, relative to base
The single passenger and driver of cooperation brush is directly determined in first set and second set, can not only reduce the data of processing data
Amount can also improve the accuracy rate for determining the single driver and passenger of cooperation brush.
Step S4 determines cooperation brush is single in first subset and the second subset first kind user and the second class
Type user.
In one embodiment, verifying model can be trained in advance by machine learning, verifying model can embody brush
The group of the group of the single driver passenger single with brush cooperates relationship when brush list, can identify the first son based on verifying model
Cooperate brush single first kind user and Second Type user in collection and the second subset, and then cooperation brush list can be exported
First kind user and Second Type user check for operator personnel analysis.
It, can be to the collection of user's composition according to the present embodiment relative to the case where individual is identified is directed in the related technology
Conjunction is identified, such as is identified to passenger's set and driver's set, is beneficial to improve recognition efficiency, and for complexity
Group's cooperation brush single game scape, which is also able to carry out, to be efficiently identified.
Fig. 2 is another schematic flow diagram for brushing single recognition methods shown in embodiment according to the present invention.Such as Fig. 2 institute
Show, on the basis of embodiment shown in Fig. 1, the method also includes:
Step S5, according to the order information of the first kind user in the first set of i-th extension, i-th extends institute
The Second Type user in second set is stated, i is positive integer；
Step S6, according to the order information of the Second Type user in the second set of i-th extension, i+1 time extension
First kind user in the first set；
Step S7, the first set to i+1 time extension and the Second Type user in the second set of i extension carry out
Cluster obtains the first subset of i+1 time cluster and the second subset of i-th cluster；
Whether the second subset of step S8, the first subset and i-th cluster that determine i+1 time cluster restrain；
Step S9, if the second subset convergence of the first subset of i+1 time cluster and i-th cluster, determines that i+1 time is poly-
Cooperation brush single first kind user and Second Type user in the second subset of the first subset and the i-th cluster of class；
Wherein, however, it is determined that cooperate brush single first kind user and Second Type user out, by the first of i+1 time cluster
I is increased 1, according to the first kind user in the first set of i-th extension by the first set that subset is extended as i-th
Order information, i-th extends the Second Type user in the second set, until the first subset of i+1 time cluster or
The second subset convergence of i-th cluster, or until can not determine cooperation brush single first kind user and Second Type user.
In one embodiment, in addition to that can be extended to first set, second set can also be extended, example
Such as after i-th extends first set, it can be believed according to the order of the first kind user in the first set that i-th extends
Breath, i-th extends the Second Type user in second set, and then can be extended again to first set, according to i-th
The order information of Second Type user in the second set of extension, i+1 time extend the first kind in the first set
User, and so on, first set and second set can be repeatedly extended, so that first set includes the first kind as much as possible
Type user, and second set is made to include Second Type user as much as possible, will pass through subsequent step as much as possible
The efficiency for identifying cooperation brush single first kind user and Second Type user, to improve recognition efficiency.
In addition, for after extending every time first set and second set can cluster, such as to i+1 time extension
First set clustered to obtain the first subset of i+1 time cluster, clustered to obtain the to the second set of i extension
The second subset of i cluster, and then the expanded set on the basis of subset of cluster advantageously ensure that in the set after extending
User is more likely the single user of cooperation brush.
It should be noted that the expansion for first set and second set can stop in some cases, such as
The second subset convergence of the first subset and i-th cluster of i+1 time cluster, or can not determine the single first kind of cooperation brush
User and Second Type user, so that it may stop continuing.
Wherein, if the second subset convergence of the first subset of i+1 time cluster and i-th cluster, illustrate first set and
Second set has had spread over the degree for meeting and needing, such as there is no the numbers for increasing user in set by extension, then
Go down without continuing extension, so as to stop extending.And if it is determined that not going out cooperation brush single first kind user and the second class
Type user illustrates that the extension to first set and second set there is a problem, if first set will likely be made by continuing extension
Include not cooperate the single users of brush more with second set, cause can not more to determine first kind user that cooperation brush is single and
Second Type user, therefore can also stop extending.
And if can determine that out and cooperate brush single first kind user and Second Type user, and the first of i+1 time cluster
The second subset of subset or i-th cluster does not restrain, then the first subset that i+1 times clusters can be extended as i-th
I is increased 1 by first set, according to the order information of the first kind user in the first set of i-th extension, i-th extension
Second Type user in the second set, to continue to extend first set and second set.
In addition, if being not determined by cooperation brush single first kind user and Second Type user, being said in the case where i=1
Bright originally determined first set is wrong, can reselect first kind user to constitute first set.
Fig. 3 is the schematic flow diagram of the single recognition methods of another brush shown in embodiment according to the present invention.Such as Fig. 3 institute
Show, on the basis of embodiment shown in Fig. 2, the first subset of the determining i+1 time cluster and the second subset of i-th cluster
Whether convergence includes:
Whether the first subset of step S801, the first subset and i-th cluster that determine i+1 time cluster are identical, and
Whether the second subset of i-th cluster and the second subset of (i-1)-th cluster are identical；
Step S802, if the first subset that the first subset of i+1 time cluster and i-th cluster is identical, and i-th clusters
The second subset second subset that is clustered with (i-1)-th time it is identical, determine the first subset and i-th cluster of i+1 time cluster
Second subset convergence.
In one embodiment, if the first subset for clustering of the first subset and i-th of i+1 time cluster is identical, namely
The first subset determined according to the first subset newly expanded and determined according to preceding the first subset once expanded
One subset is identical, and such case illustrates increasing the user in the first subset by extension, and there is no need to first
Set is extended.Correspondingly, if the second subset of i-th cluster is identical with the second subset of (i-1)-th cluster, illustrate process
Extension increases the user in second subset, and there is no need to be extended to second set.
It, can also basis in addition to the mode of embodiment according to Fig.3, judges whether the first subset and second subset restrain
It needs that other modes is selected to be judged.
If such as the first subset that the first subset of i+1 time cluster is clustered with i-th is different, can determine i+1 time
First relative complement set of the first subset that the first subset of cluster is clustered in i-th；If the second subset and i-th-of i-th cluster
The second subset of 1 cluster is different, can determine the of the second subset that the second subset of i-th cluster is clustered at (i-1)-th time
Two relative complement sets.
Wherein, relative complement set refers to, for set A and set B, the no element of the set B that set A is included
Set, i.e. relative complement set of the set A in set B.Such as first relative complement set refer to that the first subset of i+1 time cluster is wrapped
The set of the no driver of first subset of the i-th cluster contained, the second relative complement set refer to the second subset institute of i-th cluster
The set of the no passenger of the second subset for (i-1)-th cluster for including.
If identifying cooperation brush single first kind user and Second Type user, illustrate that newly determining subset is relatively previous
The new user for determining that subset has expanded new user, and expanding is cooperation brush list, therefore for previous determining son
Collection, which continues extension, can make the subset newly determined include the single user of more cooperation brushes, namely be based on the first relative complement set
When identifying the single first kind user and Second Type user of cooperation brush with the second relative complement set, it is possible to determine that first son
Collection and the second subset do not restrain, so as to continue to extend first set and second set, to increase the first subset and the
User in two subsets.
And if be based on the first relative complement set and the second relative complement set, identify not Chu cooperation brush list first kind user and
Two type of user, namely even if extending first set and second set again, the new user expanded is also not cooperation brush list
User stops extension first set and second set so as to determine first subset and second subset convergence.
Furthermore it is also possible to determine whether to restrain according to the numerical value of i, if i is larger, it is greater than default value, is illustrated pair
It is more in the number that first set and second set extend, can determine the first subset and second subset convergence, so as to after
Continuous extension first set and second set, waste of resource.
Fig. 4 is the schematic flow diagram of the single recognition methods of another brush shown in embodiment according to the present invention.Such as Fig. 4 institute
Show, it is described in the first set of i+1 time extension and the second set of i extension on the basis of embodiment shown in Fig. 2
Second Type user clusters, and the second subset of the first subset and i-th cluster that obtain i+1 time cluster includes:
Step S701, according to the characteristic information of first kind user to the first kind in the first set of i+1 time extension
Type user carries out spatial characterization, the second class in the second set extended according to the characteristic information of Second Type user to i-th
Type user carries out spatial characterization；
Step S702 determines every two first kind user by clustering algorithm in the first set of i+1 time extension
Between first distance, and i-th extension second set in determine every two Second Type user between second away from
From；
Step S703 constitutes i+1 time cluster less than the first kind user of the first pre-determined distance based on first distance
First subset constitutes the second subset of i-th cluster based on second distance less than the Second Type user of the second pre-determined distance.
In one embodiment, characteristic information can be extracted respectively for driver and passenger, wherein extracted feature letter
Breath can be one kind, be also possible to a variety of, it is preferable that can be a variety of.Wherein it is possible to using extracted characteristic information as dimension
Degree, every kind of characteristic information can carry out spatial characterization to driver and user by characteristic information as one-dimensional, by driver and
User is expressed as the form of coordinate, value, that is, feature value in coordinate.
Such as driver and passenger, extracted characteristic information includes at least one of: portrait information, path
Information, order frequency information, order subsidize information, evaluation information.Certainly, characteristic information is not limited to above-mentioned several, specifically may be used
To be configured and select as needed.
In one embodiment, it for the driver and passenger of spatial characterization, can calculate separately between driver and driver
Distance and the distance between passenger and passenger, and then it is directed to driver and passenger, it can be passed through respectively according to determining distance poly-
Class algorithm is clustered.
The distance between such as driver, first can be calculated and concentrate every two driver, then pass through no prison
Educational inspector practises and clustering, wherein used clustering algorithm can be KNN (K arest neighbors) sorting algorithm or hierarchical cluster is calculated
Method.
Wherein, characteristic information is the characteristic information with cooperation brush simple correlation, and more close characteristic information illustrates the first son
The driver of concentration brushes single characteristic and becomes apparent, and illustrates that the brush list characteristic of driver in the first subset becomes apparent namely the first subset
In driver cooperate the single mathematical expectation of probability of brush, it is bigger to cooperate the single mathematical expectation of probability of brush relative to the driver in first set.Similarly,
Passenger in second subset cooperates the single mathematical expectation of probability of brush, cooperates the single mathematical expectation of probability of brush more relative to the passenger in second set
Greatly.
Fig. 5 is the schematic flow diagram of the single recognition methods of another brush shown in embodiment according to the present invention.Such as Fig. 5 institute
Show, on the basis of embodiment shown in Fig. 2, the first subset of the determining i+1 time cluster and the second subset of i-th cluster
The first kind user and Second Type user of middle cooperation brush list include:
Step S901, based on the verifying model obtained by machine learning, in the first subset for verifying i+1 time cluster
It is single whether the user in the second subset of user and i-th cluster cooperates brush.
In one embodiment, it can determine to cooperate really from historical data and brush single first kind user and
Two type of user are as sample, to constitute sample set, and then can be obtained by machine learning based on the sample in sample set
To verifying model.
In one embodiment, the single first kind of brush can be determined in the first set of extension and the second set of extension
Type user and Second Type user, such as be determined by way of being verified under line or according to preset rules, it then will be true
Fixed first kind user and Second Type user can be passed through with constituting sample set based on the sample in sample set
Machine learning is verified model.
Wherein, brush single first kind user and Second Type user are determined in the way of preset rules, can be directed to
Part driver in first set calculates first distribution of the quantity for completing order in unit time (such as one day), for the
Two set in part passenger calculate complete order quantity within the unit time second distribution, then calculate first distribution and
The similarity of second distribution, if similarity is higher, then it is determined that this part driver and this part passenger cooperate brush list.
It should be noted that the machine learning can be supervised learning, used machine learning algorithm, which can be, is patrolled
Collect recurrence, gradient boosted tree, neural network etc..
In one embodiment, if determining the single first kind of brush in the first set of extension and the second set of extension
User and Second Type user as sample to constitute sample set, then determining the single first kind of cooperation brush based on verifying model
Type user and Second Type user, can be in first set and second set not as the first kind user of sample and
Second Type user is determined.
Corresponding with the single embodiment of recognition methods of aforementioned brush, present invention also provides the embodiments for brushing single identification device.
The brush list identification device that the embodiment of the present invention provides can be using on the server.Installation practice can pass through
Software realization can also be realized by way of hardware or software and hardware combining.Taking software implementation as an example, it anticipates as a logic
Device in justice is to be read computer program instructions corresponding in nonvolatile memory by the processor of server where it
Get what operation in memory was formed.For hardware view, as shown in fig. 6, for according to the brush list shown in the embodiment of the present invention
A kind of hardware structure diagram of server where identification device, in addition to processor shown in fig. 6, memory, network interface and it is non-easily
Except the property lost memory, the server in embodiment where device can also include generally according to the actual functional capability of the server
Other hardware repeat no more this.
Fig. 7 is a kind of schematic block diagram of brush list identification device shown in embodiment according to the present invention.Shown in the present embodiment
The single identification device of brush can be applied to the scene that at least two class users trade.Such as net about parking lot scape, one type are used
Family is driver, and another kind of user is passenger；Such as shopping at network scene, one type user are businessmans, another kind of user is to care for
Visitor.Certainly, the applicable scene of device shown in the present embodiment is not limited to above-mentioned scene, specifically can according to need and is set
It sets.Mainly the embodiment of the present invention is illustrated under net about parking lot scape below.
As shown in fig. 7, the single identification device of shown brush includes:
Gather determining module 1, for first kind user in the first set according to the single first kind user of doubtful brush
Order information, determine the second set of the Second Type user of doubtful cooperation brush list；
Set expansion module 2 extends described for the order information according to Second Type user in the second set
First kind user in one set；
Cluster module 3 obtains the first subset and the second son for clustering respectively to first set and second set
Collection；
Single determining module 4 is brushed, for determining the first kind of cooperation brush list in first subset and the second subset
User and Second Type user.
Fig. 8 is another schematic block diagram for brushing single identification device shown in embodiment according to the present invention.As shown in figure 8,
On the basis of the embodiment shown in fig. 7, first in first set that the set expansion module 2 is used to be extended according to i-th
The order information of type of user, i-th extend the Second Type user in the second set, and i is positive integer；And according to
The order information of Second Type user in the second set of i extension, i+1 time extend the first kind in the first set
Type user；
The cluster module 3 is used for the second class in the first set of i+1 time extension and the second set of i extension
Type user clusters, and obtains the first subset of i+1 time cluster and the second subset of i-th cluster；
Described device further include:
Determining module 5 is restrained, whether is the second subset that the first subset and i-th for determining i+1 time cluster cluster
Convergence；
Wherein, the list determining module 4 of brushing is received in the first subset of i+1 time cluster and the second subset of i-th cluster
In the case where holding back, the first kind that cooperation brush is single in the first subset of i+1 time cluster and the second subset of i-th cluster is determined
User and Second Type user；
Wherein, however, it is determined that cooperate brush single first kind user and Second Type user out, the first son that i+1 times is clustered
Collect the first set extended as i-th, i is increased 1, the first collection that the set expansion module 2 is used to extend according to i-th
The order information of first kind user in conjunction, i-th extend the Second Type user in the second set, until i+1
The second subset convergence of the first subset or the i-th cluster of secondary cluster, or until can not determine the single first kind of cooperation brush
User and Second Type user.
Fig. 9 is a kind of schematic block diagram of convergence determining module shown in embodiment according to the present invention.As shown in figure 9,
On the basis of Fig. 7 or embodiment illustrated in fig. 8, the convergence determining module 5 includes:
Identical determining submodule 501, for determining the first subset of i+1 time cluster and the first subset of i-th cluster
Whether the second subset of second subset and (i-1)-th cluster that whether identical and i-th clusters is identical；
Wherein, if the first subset for cluster of the first subset and i-th of i+1 time cluster is identical, and i-th cluster the
The second subset that two subsets are clustered with (i-1)-th time is identical, determines the first subset of i+1 time cluster and i-th cluster second
Subset convergence.
Figure 10 is a kind of schematic block diagram of cluster module shown in embodiment according to the present invention.As shown in Figure 10, in Fig. 7
Or on the basis of embodiment illustrated in fig. 8, the cluster module 3 includes:
Spatial characterization submodule 301, for being collected according to the characteristic information of first kind user to the first of i+1 time extension
First kind user in conjunction carries out spatial characterization, the second collection extended according to the characteristic information of Second Type user to i-th
Second Type user in conjunction carries out spatial characterization；
Distance determines submodule 302, for determining every two in the first set of i+1 time extension by clustering algorithm
First distance between first kind user, and every two Second Type user is determined in the second set of i-th extension
Between second distance；
User clusters submodule 303, constitutes for the first kind user based on first distance less than the first pre-determined distance
First subset of i+1 time cluster constitutes i-th cluster less than the Second Type user of the second pre-determined distance based on second distance
Second subset.
Figure 11 is a kind of schematic block diagram of brush list determining module shown in embodiment according to the present invention.As shown in figure 11,
On the basis of Fig. 7 or embodiment illustrated in fig. 8, the single determining module 4 of the brush includes:
Submodule 401 is verified, for based on the verifying model obtained by machine learning, verifying the of i+1 time cluster
It is single whether the user in one subset and the user in the second subset of i-th cluster cooperate brush.
The function of modules and the realization process of effect are specifically detailed in the above method and correspond to step in above-mentioned apparatus
Realization process, details are not described herein.
For device embodiment, since it corresponds essentially to embodiment of the method, so related place is referring to method reality
Apply the part explanation of example.The apparatus embodiments described above are merely exemplary, wherein described be used as separation unit
The unit of explanation may or may not be physically separated, and component shown as a unit can be or can also be with
It is not physical unit, it can it is in one place, or may be distributed over multiple network units.It can be according to actual
The purpose for needing to select some or all of the modules therein to realize application scheme.Those of ordinary skill in the art are not paying
Out in the case where creative work, it can understand and implement.
The embodiment of the present invention also proposes a kind of electronic equipment, comprising:
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to executing the step in the single recognition methods of brush described in any of the above-described embodiment.
The embodiment of the present invention also proposes a kind of computer readable storage medium, is stored thereon with computer program, the journey
The step in the single recognition methods of brush described in any of the above-described embodiment is executed when sequence is executed by processor.
The foregoing is merely the preferred embodiments of the application, not to limit the application, all essences in the application
Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the application protection.
Claims (12)
1. a kind of single recognition methods of brush characterized by comprising
According to the order information of first kind user in the first set of the single first kind user of doubtful brush, doubtful cooperation is determined
Brush the second set of single Second Type user；
According to the order information of Second Type user in the second set, the first kind extended in the first set is used
Family；
First set and second set are clustered respectively, obtain the first subset and second subset；
Determine the first kind user and Second Type user that cooperation brush is single in first subset and the second subset.
2. the method according to claim 1, wherein further include:
According to the order information of the first kind user in the first set of i-th extension, i-th extends in the second set
Second Type user, i is positive integer；
According to the order information of the Second Type user in the second set of i-th extension, the i+1 time extension first set
In first kind user；
The Second Type user in first set and the i second set extended to i+1 time extension clusters, and obtains the
The second subset of i+1 the first subset clustered and i-th cluster；
Whether the second subset of the first subset and i-th cluster that determine i+1 time cluster restrains；
If the second subset convergence of the first subset and i-th cluster of i+1 time cluster, determines the first son of i+1 time cluster
Cooperation brush single first kind user and Second Type user in the second subset of collection and i-th cluster；
Wherein, however, it is determined that cooperate brush single first kind user and Second Type user out, the first subset that i+1 times is clustered is made
For the first set of i-th extension, i is increased 1, according to the order of the first kind user in the first set of i-th extension
Information, i-th extend the Second Type user in the second set, until the first subset or i-th of i+1 time cluster are poly-
The second subset of class restrains, or until can not determine cooperation brush single first kind user and Second Type user.
3. according to the method described in claim 2, it is characterized in that, the first subset and i-th of the determining i+1 time cluster
Whether the second subset of cluster restrains
Whether the first subset of the first subset and i-th cluster that determine i+1 time cluster identical and i-th cluster the
Whether the second subset of two subsets and (i-1)-th cluster is identical；
If the first subset that the first subset and i-th of i+1 time cluster clusters is identical, and the second subset of i-th cluster with
The second subset of (i-1)-th cluster is identical, determines the second subset convergence of the first subset and i-th cluster of i+1 time cluster.
4. according to the method in claim 2 or 3, which is characterized in that described to the first set of i+1 time extension and i times
Second Type user in the second set of extension clusters, what the first subset and i-th for obtaining i+1 time cluster clustered
Second subset includes:
Space is carried out to the first kind user in the first set of i+1 time extension according to the characteristic information of first kind user
It characterizes, the Second Type user in the second set extended according to the characteristic information of Second Type user to i-th carries out space
Characterization；
By clustering algorithm determined in the first set of i+1 time extension first between every two first kind user away from
From, and the second distance between every two Second Type user is being determined in the second set of i-th extension；
The first subset for being constituted i+1 time cluster less than the first kind user of the first pre-determined distance based on first distance, is based on
Second distance constitutes the second subset of i-th cluster less than the Second Type user of the second pre-determined distance.
5. according to the method in claim 2 or 3, which is characterized in that the first subset of the determining i+1 time cluster and the
Cooperation brush single first kind user and Second Type user include: in the second subset of i cluster
Based on the verifying model obtained by machine learning, the user and i-th verified in the first subset of i+1 time cluster gathers
It is single whether the user in the second subset of class cooperates brush.
6. a kind of single identification device of brush characterized by comprising
Gather determining module, the order for first kind user in the first set according to the single first kind user of doubtful brush
Information determines the second set of the Second Type user of doubtful cooperation brush list；
Set expansion module extends first collection for the order information according to Second Type user in the second set
First kind user in conjunction；
Cluster module obtains the first subset and second subset for clustering respectively to first set and second set；
Brush single determining module, for determine cooperation brush is single in first subset and the second subset first kind user and
Second Type user.
7. device according to claim 6, which is characterized in that the set expansion module was used to be extended according to i-th
The order information of first kind user in first set, i-th extend the Second Type user in the second set, and i is
Positive integer；And the order information according to the Second Type user in the second set of i-th extension, described in i+1 time extension
First kind user in first set；
The cluster module is used to use the Second Type in the first set of i+1 time extension and the second set of i extension
Family is clustered, and the first subset of i+1 time cluster and the second subset of i-th cluster are obtained；
Described device further include:
Determining module is restrained, whether the second subset that the first subset and i-th for determining i+1 time cluster cluster restrains；
Wherein, described to brush single determining module in the first subset of i+1 time cluster and the convergent feelings of second subset of i-th cluster
Under condition, determine cooperation brush is single in the first subset of i+1 time cluster and the second subset of i-th cluster first kind user and
Second Type user；
Wherein, however, it is determined that cooperate brush single first kind user and Second Type user out, the first subset that i+1 times is clustered is made
For the first set of i-th extension, i is increased 1, in the first set that the set expansion module is used to extend according to i-th
First kind user order information, i-th extends the Second Type user in the second set, until i+1 time is poly-
The second subset convergence of the first subset or the i-th cluster of class, or until can not determine the single first kind user of cooperation brush
With Second Type user.
8. device according to claim 7, which is characterized in that the convergence determining module includes:
Identical determining module, whether the first subset that the first subset and i-th for determining i+1 time cluster cluster is identical,
And i-th cluster second subset and (i-1)-th time cluster second subset it is whether identical；
Wherein, if the first subset for cluster of the first subset and i-th of i+1 time cluster is identical, and the second of i-th cluster is sub
Collect identical with the second subset that (i-1)-th time clusters, determines the first subset of i+1 time cluster and the second subset of i-th cluster
Convergence.
9. device according to claim 7 or 8, which is characterized in that the cluster module includes:
Spatial characterization submodule, for the characteristic information according to first kind user in the first set of i+1 time extension
First kind user carries out spatial characterization, in the second set extended according to the characteristic information of Second Type user to i-th
Second Type user carries out spatial characterization；
Distance determines submodule, for determining the every two first kind in the first set of i+1 time extension by clustering algorithm
First distance between type user, and determined between every two Second Type user in the second set of i-th extension
Second distance；
User clusters submodule, constitutes i+1 time for the first kind user based on first distance less than the first pre-determined distance
First subset of cluster constitutes the second of i-th cluster less than the Second Type user of the second pre-determined distance based on second distance
Subset.
10. device according to claim 7 or 8, which is characterized in that the single determining module of the brush includes:
Submodule is verified, for verifying in the first subset that i+1 time clusters based on the verifying model obtained by machine learning
User and i-th cluster second subset in user whether to cooperate brush single.
11. a kind of electronic equipment characterized by comprising
Processor；
Memory for storage processor executable instruction；
Wherein, the processor is configured to perform claim requires any one of 1 to 5 the method.
12. 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 in any one of claims 1 to 5 the method is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810886770.9A CN109285009B (en) | 2018-08-06 | 2018-08-06 | Bill brushing identification method and bill brushing identification device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810886770.9A CN109285009B (en) | 2018-08-06 | 2018-08-06 | Bill brushing identification method and bill brushing identification device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109285009A true CN109285009A (en) | 2019-01-29 |
| CN109285009B CN109285009B (en) | 2021-03-02 |
Family
ID=65182948
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810886770.9A Active CN109285009B (en) | 2018-08-06 | 2018-08-06 | Bill brushing identification method and bill brushing identification device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109285009B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111768258A (en) * | 2019-06-05 | 2020-10-13 | 北京京东尚科信息技术有限公司 | Method, device, electronic equipment and medium for identifying abnormal order |
| CN112508630A (en) * | 2021-01-29 | 2021-03-16 | 腾讯科技（深圳）有限公司 | Abnormal conversation group detection method and device, computer equipment and storage medium |
| CN114301711A (en) * | 2021-12-31 | 2022-04-08 | 招商银行股份有限公司 | Anti-riot brushing method, device, equipment, storage medium and computer program product |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140207694A1 (en) * | 2013-01-21 | 2014-07-24 | International Foundation for Electoral Systems | Electoral integrity assessment method, and system |
| JP2016012236A (en) * | 2014-06-30 | 2016-01-21 | ぴあ株式会社 | Merchandise selling device, merchandise selling program, and merchandise selling method |
| CN105392113A (en) * | 2015-12-02 | 2016-03-09 | 百度在线网络技术（北京）有限公司 | Identification method and device for vehicle click farming |
| US20160217532A1 (en) * | 2015-01-23 | 2016-07-28 | Sure, Inc. | Securing Claim Data via Block-Chains for a Peer to Peer Platform |
| CN106127505A (en) * | 2016-06-14 | 2016-11-16 | 北京众成汇通信息技术有限公司 | The single recognition methods of a kind of brush and device |
| CN106384273A (en) * | 2016-10-08 | 2017-02-08 | 江苏通付盾科技有限公司 | Malicious order scalping detection system and method |
| CN106557955A (en) * | 2016-11-29 | 2017-04-05 | 流量海科技成都有限公司 | Net about car exception order recognition methodss and system |
| CN107093090A (en) * | 2016-10-25 | 2017-08-25 | 北京小度信息科技有限公司 | Abnormal user recognition methods and device |
| CN107872767A (en) * | 2017-11-07 | 2018-04-03 | 中国联合网络通信集团有限公司 | A kind of net about car brush single act recognition methods and identifying system |
-
2018
- 2018-08-06 CN CN201810886770.9A patent/CN109285009B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140207694A1 (en) * | 2013-01-21 | 2014-07-24 | International Foundation for Electoral Systems | Electoral integrity assessment method, and system |
| JP2016012236A (en) * | 2014-06-30 | 2016-01-21 | ぴあ株式会社 | Merchandise selling device, merchandise selling program, and merchandise selling method |
| US20160217532A1 (en) * | 2015-01-23 | 2016-07-28 | Sure, Inc. | Securing Claim Data via Block-Chains for a Peer to Peer Platform |
| CN105392113A (en) * | 2015-12-02 | 2016-03-09 | 百度在线网络技术（北京）有限公司 | Identification method and device for vehicle click farming |
| CN106127505A (en) * | 2016-06-14 | 2016-11-16 | 北京众成汇通信息技术有限公司 | The single recognition methods of a kind of brush and device |
| CN106384273A (en) * | 2016-10-08 | 2017-02-08 | 江苏通付盾科技有限公司 | Malicious order scalping detection system and method |
| CN107093090A (en) * | 2016-10-25 | 2017-08-25 | 北京小度信息科技有限公司 | Abnormal user recognition methods and device |
| CN106557955A (en) * | 2016-11-29 | 2017-04-05 | 流量海科技成都有限公司 | Net about car exception order recognition methodss and system |
| CN107872767A (en) * | 2017-11-07 | 2018-04-03 | 中国联合网络通信集团有限公司 | A kind of net about car brush single act recognition methods and identifying system |
Non-Patent Citations (2)
| Title |
|---|
| 乐元: ""基于C2C电子商务市场虚假销量识别的若干技术研究"", 《中国优秀硕士学位论文全文数据库 经济与管理科学辑》 * |
| 王忠群 等: ""基于模板用户信息搜索行为和统计分析的共谋销量欺诈识别"", 《现代图书情报技术》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111768258A (en) * | 2019-06-05 | 2020-10-13 | 北京京东尚科信息技术有限公司 | Method, device, electronic equipment and medium for identifying abnormal order |
| CN112508630A (en) * | 2021-01-29 | 2021-03-16 | 腾讯科技（深圳）有限公司 | Abnormal conversation group detection method and device, computer equipment and storage medium |
| CN114301711A (en) * | 2021-12-31 | 2022-04-08 | 招商银行股份有限公司 | Anti-riot brushing method, device, equipment, storage medium and computer program product |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109285009B (en) | 2021-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109978893B (en) | Training method, device, equipment and storage medium of image semantic segmentation network | |
| CN105574098B (en) | The generation method and device of knowledge mapping, entity control methods and device | |
| CN108595585B (en) | Sample data classification method, model training method, electronic equipment and storage medium | |
| CN108875522A (en) | Face cluster methods, devices and systems and storage medium | |
| CN109241711A (en) | User behavior recognition method and device based on prediction model | |
| CN110046706B (en) | Model generation method and device and server | |
| CN109766872A (en) | Image recognition method and device | |
| CN106951825A (en) | A kind of quality of human face image assessment system and implementation method | |
| CN110222171A (en) | A kind of application of disaggregated model, disaggregated model training method and device | |
| CN109241709A (en) | User behavior recognition method and device based on the verifying of sliding block identifying code | |
| CN109285009A (en) | It brushes single recognition methods and brushes single identification device | |
| CN110348580A (en) | Construct the method, apparatus and prediction technique, device of GBDT model | |
| CN110096617B (en) | Video classification method and device, electronic equipment and computer-readable storage medium | |
| CN111491202B (en) | Video publishing method, device, equipment and storage medium | |
| CN105814582A (en) | Method and system for recognizing faces | |
| US20230354979A1 (en) | Scalp and hair management system | |
| CN106651973A (en) | Image structuring method and device | |
| CN108717520A (en) | A kind of pedestrian recognition methods and device again | |
| CN109743286A (en) | A kind of IP type mark method and apparatus based on figure convolutional neural networks | |
| CN110458875A (en) | Detection method, image split-joint method, related device and the equipment of abnormal point pair | |
| CN109685805A (en) | A kind of image partition method and device | |
| CN110532895B (en) | Method, device and equipment for detecting fraudulent behavior in face recognition process | |
| CN109308564A (en) | The recognition methods of crowd's performance ratings, device, storage medium and computer equipment | |
| CN106372652A (en) | Hair style identification method and hair style identification apparatus | |
| Divakar et al. | Performance evaluation of credit card fraud transactions using boosting algorithms |
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 |