The content of the invention
The technical problems to be solved by the invention are to provide a kind of brush list based on equipment mark code and social group information
Detection method and system, can effectively detect brush single group and knit the brush single act in a organized way carried out by social platform software.
To solve the above problems, the present invention proposes a kind of brush list detection side based on equipment mark code and social group information
Method, comprises the following steps:
S1:Obtain on the social group information and electric business platform in social platform consumer and commodity in certain period of time
Interactive information;When including social account, the social account group, social account social connections occur for the social activity group information
Used equipment mark code;The interactive information includes interaction account, interaction commodity sign, interaction account and interaction when institute occurs
The equipment mark code used;
S2:Used equipment mark code is interacted with interacting account when social connections occur according to the social account
When used equipment mark code between association, each social account is matched with interacting account;
S3:Some transaction records to be detected of a commodity are obtained, it is corresponding to choose some interaction accounts merchandised
Social account, judges whether group relation, identical if there is some social account groups, then corresponding transaction quilt
It is easy to be just judged to brush single cross.
According to one embodiment of present invention, the step S2 includes:
Step S21:For each social account, the device identification of each social equipment of identical social account will be logged in
Mutually bound between code and corresponding social account, alternatively, the equipment by each social equipment for logging in identical social account
Mutually bind between identification code and corresponding social account and also mutually bound between the equipment mark code of each social equipment;
Step S22:For each interactive account, the device identification of each interactive device of identical interactive account will be logged in
Code between corresponding interactive account with mutually binding, alternatively, the equipment by each interactive device for logging in identical interactive account
Mutually also mutually bound between binding and the equipment mark code of each interactive device between identification code and corresponding interactive account;
Step S23:If the equipment mark code of social equipment is identical with the equipment mark code of interactive device or there is binding
Relation, then matched corresponding social account with corresponding interactive account.
According to one embodiment of present invention, the interactive account includes shopping account and/or payment accounts;
When interaction account is shopping account and payment accounts, in the step S22, also by the shopping account with paying
Account is mutually bound, and will log in the equipment mark code of each interactive device of identical interactive account and corresponding shopping account
Or mutually bound between payment accounts.
According to one embodiment of present invention, the equipment mark code includes:Phone number, cell phone apparatus IMEI code,
One or more in the MAC Address of PC, the MAC Address of tablet computer.
According to one embodiment of present invention, step S4 is further included after the step S3:
For just the easy transaction of brush single cross is judged in the step S3, its corresponding social account there are group relation is detected
It whether there is friend relation between number, if the social account in the group there are friend relation reaches pre-set ratio, be determined as
Non- brush single cross is easy, is otherwise determined as that brush single cross is easy.
According to one embodiment of present invention, the judgement of the friend relation includes:
According to nearest 1 year social association occurrence frequency γyWith nearest one month social occurrence frequency γmCalculate social account
Between social associated degree of strength β=(γy+γm);
It is otherwise non-friend relation for friend relation if social associated degree of strength β reaches predeterminable level value.
According to one embodiment of present invention, step S5 is further included after the step S4:
For the easy transaction of brush single cross is judged as in the step S4, pass through the decision tree pair based on general evaluating characteristic
Whether it is that brush single cross is easily further judged.
According to one embodiment of present invention, the construction step of the decision tree based on general evaluating characteristic includes:
A1:The easy criterion of brush single cross is set, the evaluation behavior with following characteristics is determined as that group's brush single cross is easy:
Assessment Rate is more than the first predetermined threshold value, positive rating are more than the second predetermined threshold value and first time Assessment Rate is more than the 3rd default threshold
Value;
A2:Attributive character is extracted from more brush single crosses easily and generates training dataset, the attributive character includes evaluation
Rate, positive rating, first time Assessment Rate;
A3:The information gain for each attributive character that training data concentration includes is calculated with reference to the criterion of step A1,
The decision tree based on general evaluating characteristic according to information gain selects split vertexes construction.
According to one embodiment of present invention, the step S5 includes:
S51:For the easy transaction of brush single cross is judged as in the step S4, corresponding information on commodity comment is obtained, is wrapped
Include Assessment Rate, evaluation quality, the timely degree of evaluation;
S52:According to information on commodity comment Calculation Estimation rate, positive rating, first time Assessment Rate, and led to according to result of calculation
Cross the decision tree based on general evaluating characteristic and judge whether the transaction is that brush single cross is easy.
The present invention also provides a kind of computer installation, including memory and processor and storage are on a memory and can quilt
The computer program that processor calls, when the processor performs the computer program, is realized as any in previous embodiment
The brush list detection method based on equipment mark code and social group information described in one.
The present invention also provides a kind of computer-readable recording medium, computer program is stored thereon with, the computer program
When being executed by processor, realize as described in any one in previous embodiment based on equipment mark code and social activity group information
The single detection method of brush.
The present invention also provides a kind of brush list detecting system based on equipment mark code and social group information, including:
Data obtaining module, performs on the social group information and electric business platform obtained in social platform in certain period of time
Consumer and the interactive information of commodity;The social activity group information includes social account, the social account group, social account
Used equipment mark code during generation social connections;The interactive information includes interaction account, interaction commodity sign, interaction account
Number occur interaction when used equipment mark code;
Account matching module, performs used equipment mark code when social connections occur according to the social account and hands over
Association during interaction between used equipment mark code occurs for mutual account, by each social account with interacting account progress
Match somebody with somebody;
Brush is single just to sentence module, perform some transaction records to be detected for obtaining a commodity, chooses some merchandise
The corresponding social account of interaction account, judges whether group relation, identical if there is some social account groups,
Then it is easy to be just judged to brush single cross for corresponding transaction.
According to one embodiment of present invention, the brush list detecting system based on equipment mark code and social group information
Further include:
Brush single two sentences module, perform for the brush it is single just sentence in module be just judged to the easy transaction of brush single cross, it is right to detect its
It whether there is friend relation between the social account there are group relation answered, if there are the social account of friend relation in the group
Number reach pre-set ratio, be then determined as that non-brush single cross is easy, be otherwise determined as that brush single cross is easy.
According to one embodiment of present invention, the brush list detecting system based on equipment mark code and social group information
Further include:
Brush single three sentences module, sentences for the brush single two and the easy transaction of brush single cross is judged as in module, by based on logical
Whether it is that brush single cross is easily further judged to it with the decision tree of evaluating characteristic.
After adopting the above technical scheme, the present invention has the advantages that compared with prior art:
Show accurately to screen when in face of the new list of brush in a organized way for the prior art and highly intend true brush single file
For the defects of, introduce the concept of social group, by judging that the social account corresponding to interaction account whether there is group relation,
So as to detect the brush single act largely realized by the inherent tissue and association mode between social software;
It can effectively overcome existing wash sale discrimination system in new wash sale using true logistics number, true sale
Behavior, true purchaser and addressee information cause detecting to be failed in the case of covering wash sale behavior the problem of.
Embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, below in conjunction with the accompanying drawings to the present invention
Embodiment be described in detail.
Many details are elaborated in the following description in order to fully understand the present invention.But the present invention can be with
Much implement different from other manner described here, those skilled in the art can be in the situation without prejudice to intension of the present invention
Under do similar popularization, therefore the present invention is not limited to the specific embodiments disclosed below.
Referring to Fig. 1, in one embodiment, the brush list detection method based on equipment mark code and social group information, is wrapped
Include following steps:
S1:Obtain on the social group information and electric business platform in social platform consumer and commodity in certain period of time
Interactive information;When including social account, the social account group, social account social connections occur for the social activity group information
Used equipment mark code;The interactive information includes interaction account, interaction commodity sign, interaction account and interaction when institute occurs
The equipment mark code used;
S2:Used equipment mark code is interacted with interacting account when social connections occur according to the social account
When used equipment mark code between association, each social account is matched with interacting account;
S3:Some transaction records to be detected of a commodity are obtained, it is corresponding to choose some interaction accounts merchandised
Social account, judges whether group relation, identical if there is some social account groups, then corresponding transaction quilt
It is easy to be just judged to brush single cross.
The brush list detection method based on equipment mark code and social group information of the embodiment of the present invention is carried out more below
Specific description, but should not be as limit.
In step sl, consumer in certain period of time is obtained on the social group information and electric business platform in social platform
With the interactive information of commodity.Social platform can be included multiple social platforms in the embodiment of the present invention with more than one.
Electric business platform is specifically unlimited, and commodity are certainly also unlimited, in the present invention, can be carried out every time only for a commodity under the commodity
Brush single cross easily judges, but is also not necessarily limited to this, naturally it is also possible to which multiple commodity carry out judging parallel etc..Above-mentioned certain time
Section can be set as needed, and be, for example, several days.The acquisition of these above-mentioned information can be realized by big data technology.
Social group information includes social account, the social account group, social account and occurs to be made during social connections
Equipment mark code, is certainly not limited to this.It is mechanical, electrical that the equipment that social account occurs to use during social connections can include hand
Brain, tablet etc., its corresponding equipment mark code is with can including the MAC of phone number, cell phone apparatus IMEI code, PC
Location, the MAC Address etc. of tablet computer.Social platform can be YY platforms, QT platforms, QQ, wechat etc., and group can be voice
Group, QQ groups, wechat group etc..
The interactive information includes interaction account, interaction commodity sign, interaction account and used equipment during interaction occurs
Identification code.Certainly, interactive information is certainly not limited to this, and the interactive information that user produces on electric business platform, can browse for user
Commodity, the information bought commodity and submitted when being paid to respective transaction, interactive information can include the purchase of shopping platform
Thing account, the SKU numberings (stock number, can identify as tradable commodity) for browsing commodity, the SKU volumes for buying commodity
Number, the payment accounts of payment platform, interaction when the equipment that uses and corresponding equipment mark code.Equipment can for example include hand
Mechanical, electrical brain, tablet etc., corresponding equipment mark code can include phone number, cell phone apparatus IMEI code, the MAC of PC
Address, the MAC Address of tablet computer.Electric business platform is, for example, that Taobao, Jingdone district, poly- U.S. etc. are unlimited.Interactive information can include
Transaction record information caused by consumer transaction commodity, can also include consumer and browse the goods browse that commodity are generations
Information etc..
After the acquisition of information of needs is collected, step S2 is then performed, referring to Fig. 5, is occurred according to the social account
Association when used equipment mark code is with interacting account generation interaction during social connections between used equipment mark code,
Each social account is matched with interacting account.
Optionally, equipment mark code includes:Phone number, the IMEI code of cell phone apparatus, the MAC Address, flat of PC
One or more in the MAC Address of plate computer.Usually, user may possess different equipment or the equipment more renewed
Use, thus identical social account or interaction account can be logged in different equipment, have multiple social activities in the case
Equipment mark code and the corresponding interactive account of the equipment mark code of equipment and corresponding social account relating and multiple interactive devices
Number association, these equipment mark codes can be bound with corresponding account or these equipment mark codes between can also
Mutually binding.In an only account associate device identification code, then both are bound.Since no matter which is logged in is flat
Platform, the equipment mark code of equipment will not become, thus social account and interaction account can be identified by equipment mark code
Whether it is same user, so that both be matched.
Further, step S2 comprises the following steps:
Step S21:For each social account, the device identification of each social equipment of identical social account will be logged in
Mutually bound between code and corresponding social account, alternatively, the equipment by each social equipment for logging in identical social account
Mutually bind between identification code and corresponding social account and also mutually bound between the equipment mark code of each social equipment.
The social equipment of identical social account, including mobile phone, PC, tablet computer etc. were logged in, for associated equipment, such as
User logged in same QQ accounts with oneself mobile phone and computer and tablet computer, then was wrapped in the social information got
Account containing QQ, the phone number of user, the IMEI code of user mobile phone, the MAC Address of user computer, the MAC Address of tablet computer.
By the phone number of user, the IMEI code of user mobile phone, the MAC Address of user computer, the MAC Address of tablet computer and the QQ accounts
Number binding.
Step S22:For each interactive account, the device identification of each interactive device of identical interactive account will be logged in
Code between corresponding interactive account with mutually binding, alternatively, the equipment by each interactive device for logging in identical interactive account
Mutually also mutually bound between binding and the equipment mark code of each interactive device between identification code and corresponding interactive account.
Equipment such as mobile phone, computer, the tablet computer of identical interactive account were logged in, for associated equipment.Such as user uses oneself
Mobile phone and computer and tablet computer logged in same shopping account (such as Taobao's account, Jingdone district account etc.), then be
Phone number comprising user in the interactive information that gets of uniting, shopping account, the IMEI code of user mobile phone, user computer
The MAC Address of MAC Address, tablet computer.
Optionally, interaction account includes shopping account and/or payment accounts.Pay a bill in interaction account for shopping account or branch
Number when, as long as the binding of device flag code and corresponding account is got up.It is shopping account and payment accounts in interaction account
When, in the step S22, also the shopping account is mutually bound with payment accounts, each of identical interactive account will be logged in
Mutually bound between the equipment mark code of a interactive device and corresponding shopping account or payment accounts.In other words, if getting
Interactive information in further include payment accounts (such as Alipay account), then shopping account and payment accounts are bound, and will log in
The equipment mark code crossed the equipment mark code of the equipment of same payment accounts and logged in the equipment of shopping account is bound.
Step S23:If the equipment mark code of social equipment is identical with the equipment mark code of interactive device or there is binding
Relation, then matched corresponding social account with corresponding interactive account.In other words, the device identification of social equipment is detected
The equipment mark code of code and interactive device, if the equipment mark code of social equipment is identical with the equipment mark code of interactive device or deposits
, then will corresponding social account and the corresponding matching of corresponding interaction account in binding relationship.There are binding relationship to refer to, both
Between be directly bound, or both and to be bound with same account.
The step for, interactive information and social information based on user, by between account and the equipment mark code of equipment
Incidence relation, have found social account corresponding with interaction account.
Then step S3 is performed, some transaction records (purchaser record) to be detected of a commodity is obtained, can be nearest
All or part of transaction record in a period of time.From the interaction account of these transaction records, some merchandise is chosen
Interaction account, can be all or part of interactive account merchandised, and may thereby determine that its corresponding social account, after
And judge to whether there is group relation between these social accounts by social group information, if there is some social account institutes
Identical in group, then it is easy to be just judged to brush single cross for corresponding transaction.The behavior is brush of the Shua Dan companies by social group organization
Single act.
In one embodiment, step S4 is further included afterwards referring to Fig. 2, step S3:
For just the easy transaction of brush single cross is judged in the step S3, its corresponding social account there are group relation is detected
It whether there is friend relation between number, if the social account in the group there are friend relation reaches pre-set ratio, be determined as
Non- brush single cross is easy, is otherwise determined as that brush single cross is easy.
It will can tentatively be judged as that the single transaction of brush carries out propagating the true purchase brought with social networks again in step S3
Behavior is screened.
Preferably, the judgement of friend relation includes:
According to nearest 1 year social association occurrence frequency γyWith nearest one month social occurrence frequency γmCalculate social account
Between social associated degree of strength β=(γy+γm);
It is otherwise non-friend relation for friend relation if social associated degree of strength β reaches predeterminable level value.
Step S4 can be specifically, respectively with parameter γyThe frequency that association in nearest 1 year occurs is defined, γmTo most
The frequency that nearly January, association occurred is defined.From the social group information in the social software got according to code of points pair
Parameter γy、γmCarry out value, the corresponding code of points of two parameters can be such as following table (1), but without limitation.
Table (1)
Social associated degree of strength β is calculated as follows between user:
β=(γy+γm) (1)
According to preset rules and the value of combination β account with the presence or absence of friend relation distinguishing, wherein default
Rule can be as follows:
If the account sum there are group relation is m, wherein the account number there are at least one friend relation is n, exist
The ratio of general ledger number in the account Shuo Zhan groups of friend relation:
In one embodiment, step S5 is further included afterwards referring to Fig. 3, the step S4:
For the easy transaction of brush single cross is judged as in the step S4, pass through the decision tree pair based on general evaluating characteristic
Whether it is that brush single cross is easily further judged.
The more transactions single for being judged to brush in step s 4, then the decision tree based on general evaluating characteristic to its whether
It is that brush is single further to be judged, general evaluating characteristic refers to the evaluating characteristic regardless of user, for the conventional of each user
Evaluating data can be as the data source of feature.
Preferably, the construction step of the decision tree based on general evaluating characteristic includes:
A1:The easy criterion of brush single cross is set, the evaluation behavior with following characteristics is determined as that group's brush single cross is easy:
Assessment Rate is more than the first predetermined threshold value, positive rating are more than the second predetermined threshold value and first time Assessment Rate is more than the 3rd default threshold
Value;Assessment Rate is high, favorable comment is more and evaluation can be configured in time, and features described above can be divided by the rule of setting
Determine.
A2:Attributive character is extracted from more brush single crosses easily and generates training dataset, the attributive character includes evaluation
Rate, positive rating, first time Assessment Rate;Assessment Rate is to evaluate the ratio of stroke count and conclusion of the business stroke count, and positive rating provides favorable comment
The ratio of stroke count and overall merit stroke count, first time Assessment Rate provide the stroke count of evaluation and the ratio of overall merit stroke count at the first time
Rate, can set as needed at the first time.
A3:The information gain for each attributive character that training data concentration includes is calculated with reference to the criterion of step A1,
The decision tree based on general evaluating characteristic according to information gain selects split vertexes construction.Information gain can be special to attribute
Sign progress is certain to be quantified, and can be gained knowledge by information to calculate information gain, details are not described herein.Structure of the embodiment of the present invention
Into decision tree it is as shown in Figure 2.
Preferably, step S5 includes:
S51:For the easy transaction of brush single cross is judged as in the step S4, corresponding information on commodity comment is obtained, is wrapped
Include Assessment Rate, evaluation quality, the timely degree of evaluation;
S52:According to information on commodity comment given price rate, positive rating, first time Assessment Rate, and passed through according to result of calculation
The decision tree based on general evaluating characteristic judges whether the transaction is that brush single cross is easy.
According in a certain commodity of the businessman got the relevant information on commodity comment of account, including evaluation are interacted with described
Rate, evaluation quality, the timely degree of evaluation, it is brush single act or social group to carry out judgement to above-mentioned transaction by the decision tree
Propagate buying behavior (i.e. non-brush single act).
The present invention is explained further with reference to a specific embodiment:
Get the corresponding history social information of 10 accounts such as following table (2) on YY voice platforms:
Table (2)
By the equipment that same YY voices account was logged in table (2) and equipment mark code, (same a line i.e. in table is set
Standby and equipment mark code) bind mutually and bound with YY voices account.
Get the historical transactional information of upper 100 users of Taobao, the equipment that 100 users were logged in into its Taobao's account
And equipment mark code binding and is bound mutually with Taobao account, is illustrated below by taking wherein 10 users as an example, 10 users
History mutual information such as following table (3):
Table (3)
By logged in table (3) same Taobao's account equipment and equipment mark code (i.e. in table with a line equipment and
Equipment mark code) bind mutually and bound with corresponding Taobao's account.
Certain double intraday 10 transaction of shoes records on some shop in acquisition Taobao, details such as following table (4):
Table (4)
1., for the first transaction:It is tablet computer using equipment, and MAC Address is 3C:D0:F8:2C:99:A2,
Equipment mark code in look-up table (3) with MAC Address binding is phone number 13205812397 and mobile phone IMEI code
354717046702222, lookup and 3C in table (2):D0:F8:2C:99:A2 and/or 13205812397 and/or
354717046702222 identical or binding equipment mark codes, find the equipment mark code of 2 that a line of serial number in table (2),
Including mobile phone IMEI code:354717046702222 and tablet computer MAC Address:3C:D0:F8:2C:99:A2, its YY voice account
Number it islisi@qq.com。
2., for the second transaction, its equipment used is computer, and the MAC Address of computer is 00-01-6C-5D-46-
93, the equipment mark code in look-up table (3) with MAC Address binding is phone number:13100002345 and mobile phone IMEI code:
866723019683291, search in the table (2) with 00-01-6C-5D-46-93 and/or 13100002345 and/or
866723019683291 identical or binding equipment mark codes, find the equipment mark code of 1 that a line of serial number in table (2), bag
Include phone number:13100002345 and mobile phone IMEI code 866723019683291 and computer MAC Address 00-01-6C-5D-
46-93, its YY voice account arezhangsan@qq.com。
3., for the 3rd transaction, its equipment used is computer, and the MAC Address of its computer is 00-0C-29-58-
99-11, look-up table do not have other equipment identification code and the address binding in (3), lookup and 00-0C-29-58- in table (2)
99-11 is identical or the equipment mark code of binding, finds the equipment mark code of 4 that a line of serial number in table (2), including computer MAC
Address:00-0C-29-58-99-11 and tablet computer MAC Address 4E:33:4B:33:AB:6H, its YY voice account arexiaohong@126.com。
4., for the 4th transaction, its equipment used is mobile phone, its corresponding equipment mark code includes cell-phone number:
15821052670 and mobile phone IMEI code:860859027401111, bound with the cell-phone number and mobile phone IMEI code in look-up table (3)
Equipment mark code be 00-19-21-C4-BA-DB, in look-up table (2) with cell-phone number:15821052670 and mobile phone IMEI code:
860859027401111 and MAC Address 00-19-21-C4-BA-DB it is identical or binding equipment mark code, find serial number
Equipment mark code in 10 that a line, including mobile phone IMEI code 860859027401111 and computer MAC Address 00-19-21-
C4-BA-DB, its corresponding YY voices account areAbc1256@126.com。
5. for the 5th transaction, its equipment used is tablet computer, and MAC Address is C8:33:4B:33:AB:D7,
Equipment mark code in look-up table (3) with MAC Address binding is cell-phone number 13958013467 and mobile phone IMEI code
012425003304271, in look-up table (2) with C8:33:4B:33:AB:D7 and/or 012425003304271 is identical or binds
Equipment mark code, find that a line of serial number 3 in table (2), its corresponding YY voices account iswangwu@163.com。
6. for the 6th transaction, its equipment used is that mobile phone its mobile phone IMEI code is 863664000059021, is searched
The equipment mark code bound in table (3) with it, is the MAC Address 5B of tablet computer:3D:11:CD:5E:6C.In look-up table (2)
With 863664000059021 and/or 5B:3D:11:CD:5E:6C it is identical or binding equipment mark code, find sequence number in table (2)
For 6 that a line, its corresponding YY voices account isxiaoming@qq.com。
7. similarly, for the 7th, 8,9,10 transactions, the corresponding YY voices account of the 7th transaction is 23657890, the 8th
Corresponding YY voices account of merchandising isxiaohua@163.com, the corresponding YY voices account of the 9th transaction for 78459944, the
The corresponding YY voices account of 10 transactions is 654987321.
The YY voice accounts found according to above-mentioned steps, with reference to the group information where YY voice accounts in table (2), find
Group relation existing for each YY voices account.
Group in table (2) is taken into intersection, is foundzhangsan@qq.com、wangwu@163.com、xiaohong@ 126.com、23657890、xiaoming@qq.com、78459944、xiaohua@163.comThis 7 YY voices accounts exist
Group relation, then it is the 1st, 3,4,5,6,7,9 this 7 transaction preliminary judgement are single for brush.
Consider further that there are whether there is friend relation, the embodiment of the present invention between 7 YY voice accounts of this of group relation
In, can be between comprehensive descision its user by the information that is contacted between the information and YY voice accounts that are contacted between phone number
It is no that there are friend relation, specifying information such as following table (5):
Table (5)
Based on upper table (5) and β formula are calculated, calculate social associated degree of strength between user:
βAB=0.2;βBC=1.3;βCD=0.2;βDE=0.1;βEF=0.4;βFG=1.3;βAG=0.2 ... wherein, βAB
Social strength of association between user A and user B, βACSocial strength of association between user A and user C, and so on.
And according toJudge, be good friend between B and C, be good friend between F and G,
Each friend relation is calculated successively.The ratio there are general ledger number in the account Shuo Zhan groups of friend relation is calculated afterwardsSuch as after being calculated by friend relation, n 8, m 10, then p is 0.8.
So as to judge that the transaction is easy for brush single cross.
To be determined as in above-mentioned steps the single transaction of brush again by the decision tree based on general evaluating characteristic to its whether be
Brush is single further to be judged, is comprised the following steps that:Obtain and be judged to the single transaction of brush, if judge above in that step
It is single for doubtful brush to go out 5 transactions, then the evaluation record information of this 5 transaction is obtained, such as following table (6):
Table (6)
By setting specific Assessment Rate threshold value such as 0.8, positive rating threshold value 0.9, first time Assessment Rate threshold value 0.5,
At the first time can be set as 3 it is small when within because brush it is single as can all provide evaluation at the right time.According to the table got
(6) information is gone on decision tree again, then can be judged to the single transaction of doubtful brush by above-mentioned 5 carries out last brush list
Also it is the single judgement of non-brush.
The present invention also provides a kind of computer installation, including memory and processor and storage are on a memory and can quilt
The computer program that processor calls, when the processor performs the computer program, is realized as any in previous embodiment
The brush list detection method based on equipment mark code and social group information described in one.
The present invention also provides a kind of computer-readable recording medium, computer program is stored thereon with, the computer program
When being executed by processor, realize as described in any one in previous embodiment based on equipment mark code and social activity group information
The single detection method of brush.
The present invention also provides a kind of brush list detecting system based on equipment mark code and social group information, including:
Data obtaining module, performs on the social group information and electric business platform obtained in social platform in certain period of time
Consumer and the interactive information of commodity;The social activity group information includes social account, the social account group, social account
Used equipment mark code during generation social connections;The interactive information includes interaction account, interaction commodity sign, interaction account
Number occur interaction when used equipment mark code;
Account matching module, performs used equipment mark code when social connections occur according to the social account and hands over
Association during interaction between used equipment mark code occurs for mutual account, by each social account with interacting account progress
Match somebody with somebody;
Brush is single just to sentence module, perform some transaction records to be detected for obtaining a commodity, chooses some merchandise
The corresponding social account of interaction account, judges whether group relation, identical if there is some social account groups,
Then it is easy to be just judged to brush single cross for corresponding transaction.
According to one embodiment of present invention, the brush list detecting system based on equipment mark code and social group information
Further include:
Brush single two sentences module, perform for the brush it is single just sentence in module be just judged to the easy transaction of brush single cross, it is right to detect its
It whether there is friend relation between the social account there are group relation answered, if there are the social account of friend relation in the group
Number reach pre-set ratio, be then determined as that non-brush single cross is easy, be otherwise determined as that brush single cross is easy.
According to one embodiment of present invention, the brush list detecting system based on equipment mark code and social group information
Further include:
Brush single three sentences module, sentences for the brush single two and the easy transaction of brush single cross is judged as in module, by based on logical
Whether it is that brush single cross is easily further judged to it with the decision tree of evaluating characteristic.
Particular content on the single detecting system of the brush based on equipment mark code and social group information of the present invention can be with
Referring to the description of the single detection method part of brush based on equipment mark code and social group information in previous embodiment, herein no longer
Repeat.
The embodiment of the present invention can be by obtaining Transaction Information and social information, based on computer end software (WEB page etc.)
Or mobile phone terminal software distinct device it is identical/login record information at different moments matched, by using setting for equipment
Standby identification code finds the correspondence between interactive account and social account, then judge social activity account corresponding with interaction account it
Between whether there is group relation, so as to detect the brush largely realized by the inherent tissue and association mode between social software
Single act.
The embodiment of the present invention can effectively overcome recognition methods of the existing wash sale discrimination system using conventional brush list feature
Can not screen wash sale industrial chain using true logistics number, true sale behavior, true purchaser and addressee information into
Capable height intends the defects of true feature brush list is movable, can effectively detect brush single group and knits and is carried out in a organized way by social platform software
Brush single act.
Although the present invention is disclosed as above with preferred embodiment, it is not for limiting claim, any this area
Technical staff without departing from the spirit and scope of the present invention, can make possible variation and modification, therefore the present invention
Protection domain should be subject to the scope that the claims in the present invention are defined.