Disclosure of Invention
The invention aims to provide a method and a system for detecting a list brushing based on an equipment identification code and social group information, which can effectively detect organized list brushing actions of a list brushing organization through social platform software.
In order to solve the above problems, the present invention provides a method for detecting a swipe ticket based on an equipment identifier and social group information, comprising the following steps:
s1: acquiring social group information on a social platform and interaction information of consumers and commodities on an e-commerce platform within a certain time period; the social group information comprises a social account number, a group where the social account number is located and an equipment identification code used when the social account number is in social connection; the interactive information comprises an interactive account, an interactive commodity identifier and a device identifier used when the interactive account is interacted;
s2: matching each social account with the interactive account according to the association between the equipment identification code used when the social account is in social contact and the equipment identification code used when the interactive account is in interaction;
s3: the method comprises the steps of obtaining a plurality of transaction records to be detected of a commodity, selecting social account numbers corresponding to a plurality of interactive account numbers with transactions, judging whether a group relationship exists, and if the groups of the social account numbers are the same, judging the corresponding transactions as bill-swiping transactions.
According to an embodiment of the present invention, the step S2 includes:
step S21: for each social account, mutually binding the equipment identification code of each social device logged in the same social account with the corresponding social account, or mutually binding the equipment identification code of each social device logged in the same social account with the corresponding social account, and mutually binding the equipment identification codes of each social device;
step S22: for each interactive account, mutually binding the equipment identification code of each interactive equipment logged in the same interactive account with the corresponding interactive account, or mutually binding the equipment identification code of each interactive equipment logged in the same interactive account with the corresponding interactive account, and mutually binding the equipment identification codes of each interactive equipment;
step S23: and if the equipment identification code of the social equipment is the same as the equipment identification code of the interactive equipment or a binding relationship exists between the equipment identification code of the social equipment and the equipment identification code of the interactive equipment, matching the corresponding social account with the corresponding interactive account.
According to one embodiment of the invention, the interactive account includes a shopping account and/or a payment account;
when the interactive account is a shopping account and a payment account, in step S22, the shopping account and the payment account are bound to each other, and the device identification code of each interactive device that has logged in the same interactive account is bound to the corresponding shopping account or payment account.
According to one embodiment of the invention, the device identification code comprises: one or more of a mobile phone number, an IMEI code of mobile phone equipment, an MAC address of a personal computer and an MAC address of a tablet personal computer.
According to an embodiment of the present invention, the step S3 is followed by the step S4:
and detecting whether a friend relationship exists between the corresponding social account numbers with the group relationship aiming at the transaction which is judged to be the order-swiping transaction in the step S3, if the social account numbers with the friend relationship in the group reach a preset ratio, judging the transaction to be a non-order-swiping transaction, and otherwise, judging the transaction to be an order-swiping transaction.
According to an embodiment of the present invention, the determining of the friend relationship includes:
according to the occurrence frequency gamma of social associations in the last yearyAnd the frequency of social occurrences of the last month gammamCalculating the strength of social associations between social accounts, β ═ (γ)y+γm);
If the strength degree beta of the social association reaches a preset degree value, the relationship is a friend relationship, otherwise, the relationship is a non-friend relationship.
According to an embodiment of the present invention, the step S4 is followed by the step S5:
for the transaction determined as the swizzle transaction in the step S4, a further determination is made as to whether it is a swizzle transaction through a decision tree based on common evaluation features.
According to one embodiment of the invention, the step of constructing the decision tree based on the general evaluation features comprises:
a1: setting a judgment standard of the bill-swiping transaction, and judging the evaluation behavior with the following characteristics as a group bill-swiping transaction: the evaluation rate exceeds a first preset threshold, the good evaluation rate exceeds a second preset threshold, and the first time evaluation rate exceeds a third preset threshold;
a2: extracting attribute features from multiple brush transaction and generating a training data set, wherein the attribute features comprise an evaluation rate, a goodness evaluation rate and a first-time evaluation rate;
a3: and B, calculating the information gain of each attribute feature contained in the training data set by combining the judgment standard of the step A1, and selecting a split node according to the information gain to construct the decision tree based on the general evaluation feature.
According to an embodiment of the present invention, the step S5 includes:
s51: for the transaction determined as the transaction of the order-swiped transaction in the step S4, acquiring corresponding commodity evaluation information including an evaluation rate, evaluation quality and evaluation timeliness;
s52: and calculating the evaluation rate, the good evaluation rate and the first-time evaluation rate according to the commodity evaluation information, and judging whether the transaction is a bill-swiping transaction or not through the decision tree based on the general evaluation features according to the calculation result.
The invention also provides a computer device, which comprises a memory, a processor and a computer program stored on the memory and capable of being called by the processor, wherein when the processor executes the computer program, the method for detecting a swipe bill based on the device identification code and the social group information is implemented as described in any one of the foregoing embodiments.
The present invention also provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method for detecting a swipe form based on a device identifier and social group information as described in any one of the foregoing embodiments.
The invention also provides a system for detecting the bill swiping based on the equipment identification code and the social group information, which comprises the following steps:
the information acquisition module is used for acquiring social group information on the social platform and interaction information of consumers and commodities in a certain time period on the e-commerce platform; the social group information comprises a social account number, a group where the social account number is located and an equipment identification code used when the social account number is in social connection; the interactive information comprises an interactive account, an interactive commodity identifier and a device identifier used when the interactive account is interacted;
the account matching module is used for executing association between the equipment identification code used when the social contact occurs to the social account and the equipment identification code used when the interactive account interacts, and matching each social account with the interactive account;
the bill-swiping initial judgment module is used for acquiring a plurality of transaction records to be detected of a commodity, selecting a plurality of social account numbers corresponding to interactive account numbers with transactions, judging whether a group relationship exists or not, and if the groups of the social account numbers are the same, judging the corresponding transactions as bill-swiping transactions.
According to an embodiment of the present invention, the system for detecting a swipe form based on the device identification code and the social group information further includes:
and the bill-swiping secondary judgment module executes the transaction which is judged as the bill-swiping transaction in the bill-swiping primary judgment module, detects whether a friend relationship exists between the corresponding social account numbers with the group relationship, judges that the bill-swiping transaction is not the bill-swiping transaction if the social account numbers with the friend relationship in the group reach a preset ratio, and judges that the bill-swiping transaction is the bill-swiping transaction if the social account numbers with the friend relationship in the group do not reach the preset ratio.
According to an embodiment of the present invention, the system for detecting a swipe form based on the device identification code and the social group information further includes:
and the list-swiping three-judgment module is used for further judging whether the transaction is the list-swiping transaction or not through a decision tree based on the general evaluation features aiming at the transaction which is judged to be the list-swiping transaction in the list-swiping two-judgment module.
After the technical scheme is adopted, compared with the prior art, the invention has the following beneficial effects:
aiming at the defect that high-fidelity bill-swiping behaviors cannot be accurately discriminated when a novel organized bill-swiping mode is faced in the prior art, the concept of a social group is introduced, and a large number of single-swiping behaviors realized through an internal organization and an association mode between social software are detected by judging whether a group relation exists in a social account corresponding to an interactive account;
the problem of detection failure caused by the fact that the existing false transaction screening system covers false transaction behaviors by adopting real logistics numbers, real transaction behaviors, real purchaser and recipient information in novel false transactions can be effectively solved.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather construed as limited to the embodiments set forth herein.
Referring to fig. 1, in one embodiment, a method for detecting a swipe bill based on a device identification code and social group information includes the following steps:
s1: acquiring social group information on a social platform and interaction information of consumers and commodities on an e-commerce platform within a certain time period; the social group information comprises a social account number, a group where the social account number is located and an equipment identification code used when the social account number is in social connection; the interactive information comprises an interactive account, an interactive commodity identifier and a device identifier used when the interactive account is interacted;
s2: matching each social account with the interactive account according to the association between the equipment identification code used when the social account is in social contact and the equipment identification code used when the interactive account is in interaction;
s3: the method comprises the steps of obtaining a plurality of transaction records to be detected of a commodity, selecting social account numbers corresponding to a plurality of interactive account numbers with transactions, judging whether a group relationship exists, and if the groups of the social account numbers are the same, judging the corresponding transactions as bill-swiping transactions.
The following describes a method for detecting a swipe form based on a device identifier and social group information in an embodiment of the present invention in more detail, but the present invention should not be limited thereto.
In step S1, social group information on the social platform and interaction information between the consumer and the merchandise within a certain time period on the e-commerce platform are obtained. There may be more than one social platform, and multiple social platforms may be incorporated into embodiments of the present invention. The e-commerce platform is not limited to a specific commodity, and certainly not limited to a commodity, and in the present invention, the transaction determination of the brush list under the commodity can be performed only for one commodity at a time, but is not limited to this, and certainly, the determination can be performed in parallel for a plurality of commodities, and so on. The above-mentioned certain period of time may be set as needed, for example, several days. The above-mentioned acquisition of such information can be achieved by big data technology.
The social group information includes a social account, a group in which the social account is located, and a device identifier used when the social account is socially connected, but is not limited thereto. The device used when the social account is socially connected may include a mobile phone, a computer, a tablet computer, and the like, and the corresponding device identification code may include a mobile phone number, an IMEI code of the mobile phone device, an MAC address of the personal computer, an MAC address of the tablet computer, and the like. The social platform may be a YY platform, a QT platform, a QQ, a wechat, etc., and the group may be a voice group, a QQ group, a wechat group, etc.
The interactive information comprises an interactive account, an interactive commodity identifier and an equipment identifier used when the interactive account is interacted. Certainly, the interactive information is not limited to this, and the interactive information generated by the user on the e-commerce platform may be information submitted when the user browses goods, purchases goods, and pays for a corresponding transaction, and the interactive information may include a shopping account number of the shopping platform, a SKU number (stock number, which may be a transaction goods identifier) of the browsed goods, a SKU number of the purchased goods, a payment account number of the payment platform, equipment used during interaction, and a corresponding equipment identifier. The device may include, for example, a mobile phone, a computer, a tablet, etc., and the corresponding device identification code may include a mobile phone number, an IMEI code of the mobile phone device, a MAC address of a personal computer, and a MAC address of the tablet. The e-commerce platform is not limited to Taobao, Jingdong, Jumei, etc. The interactive information may include transaction record information generated when the consumer transacts the commodity, and may also include commodity browsing information generated when the consumer browses the commodity.
After acquiring and collecting the required information, step S2 is executed, and referring to fig. 5, each social account and the interactive account are matched according to the association between the device identifier used when the social account is in social contact and the device identifier used when the interactive account is in interaction.
Optionally, the device identification code includes: one or more of a mobile phone number, an IMEI code of mobile phone equipment, an MAC address of a personal computer and an MAC address of a tablet personal computer. Generally speaking, a user may have different devices or use new devices instead, and the like, and therefore may log in the same social account or interactive account using different devices, in this case, device identifiers of a plurality of social devices are associated with corresponding social accounts, device identifiers of a plurality of interactive devices are associated with corresponding interactive accounts, and the device identifiers may be all bound with corresponding accounts or may be bound with each other. When only one account is only associated with the equipment identification code, the account and the equipment identification code are bound. Because the device identification code of the device is unchangeable no matter which platform is logged in, whether the social account and the interactive account are the same user can be identified through the device identification code, and therefore the social account and the interactive account are matched.
Further, step S2 includes the following steps:
step S21: and aiming at each social account, binding the equipment identification code of each social device logged in the same social account with the corresponding social account, or binding the equipment identification code of each social device logged in the same social account with the corresponding social account, and binding the equipment identification codes of the social devices with each other. The social equipment which logs in the same social account number comprises a mobile phone, a personal computer, a tablet personal computer and the like, and for the associated equipment, for example, if a user logs in the same QQ account number by using the mobile phone, the personal computer and the tablet personal computer of the user, the obtained social information comprises the QQ account number, the mobile phone number of the user, the IMEI code of the mobile phone of the user, the MAC address of the computer of the user and the MAC address of the tablet personal computer. And binding the mobile phone number of the user, the IMEI code of the mobile phone of the user, the MAC address of the user computer and the MAC address of the tablet personal computer with the QQ account.
Step S22: and aiming at each interactive account, mutually binding the equipment identification code of each interactive equipment logged in the same interactive account with the corresponding interactive account, or mutually binding the equipment identification code of each interactive equipment logged in the same interactive account with the corresponding interactive account, and mutually binding the equipment identification codes of each interactive equipment. The devices, such as the mobile phone, the computer and the tablet computer, which have logged in the same interactive account are associated devices. For example, the user logs in the same shopping account (e.g., a panning account, a kyoto account, etc.) through his/her mobile phone, computer, and tablet computer, the interactive information acquired by the system includes the user's mobile phone number, the shopping account, the IMEI code of the user's mobile phone, the MAC address of the user's computer, and the MAC address of the tablet computer.
Optionally, the interactive account includes a shopping account and/or a payment account. When the interactive account is a shopping account or a payment account, the equipment mark code and the corresponding account are bound together. When the interactive account is a shopping account and a payment account, in step S22, the shopping account and the payment account are further bound to each other, and the device identification code of each interactive device that has logged in the same interactive account is bound to the corresponding shopping account or payment account. In other words, if the acquired interaction information further includes a payment account (e.g., a paypal account), the shopping account and the payment account are bound, and the device identifier of the device that has logged in the same payment account is bound with the device identifier of the device that has logged in the shopping account.
Step S23: and if the equipment identification code of the social equipment is the same as the equipment identification code of the interactive equipment or a binding relationship exists between the equipment identification code of the social equipment and the equipment identification code of the interactive equipment, matching the corresponding social account with the corresponding interactive account. In other words, the device identification code of the social device and the device identification code of the interactive device are detected, and if the device identification code of the social device and the device identification code of the interactive device are the same or have a binding relationship, the corresponding social account and the corresponding interactive account are correspondingly matched. The existence of the binding relationship means that the two are directly bound or bound with the same account.
In the step, based on the interactive information and the social information of the user, the social account corresponding to the interactive account is found out through the association relationship between the account and the equipment identification code of the equipment.
Then, step S3 is executed to obtain a plurality of transaction records (purchase records) to be detected for a commodity, which may be all or part of the transaction records in the latest period of time. Selecting a plurality of interactive account numbers with transactions from the interactive account numbers of the transaction records, wherein the interactive account numbers can be all or part of the interactive account numbers with transactions, so that the corresponding social account numbers can be determined, then judging whether group relations exist among the social account numbers through social group information, and if the groups of the social account numbers are the same, the corresponding transactions are primarily judged to be the transaction of the bill swiping. The behavior is a swipe behavior organized by social groups by a swipe company.
In one embodiment, referring to fig. 2, step S3 is followed by step S4:
and detecting whether a friend relationship exists between the corresponding social account numbers with the group relationship aiming at the transaction which is judged to be the order-swiping transaction in the step S3, if the social account numbers with the friend relationship in the group reach a preset ratio, judging the transaction to be a non-order-swiping transaction, and otherwise, judging the transaction to be an order-swiping transaction.
The transaction preliminarily determined in step S3 as a ticket may be screened from the true purchasing behavior brought by social network propagation.
Preferably, the judgment of the friend relationship comprises:
according to the occurrence frequency gamma of social associations in the last yearyAnd the frequency of social occurrences of the last month gammamCalculating the strength of social associations between social accounts, β ═ (γ)y+γm);
If the strength degree beta of the social association reaches a preset degree value, the relationship is a friend relationship, otherwise, the relationship is a non-friend relationship.
In step S4, the parameter γ may be usedyDefining the frequency of the occurrence of the last year of the association, gammamThe frequency with which the last month association occurred is defined. Parameter gamma is matched according to scoring rules from the obtained social group information on the social softwarey、γmTaking values, the scoring rules corresponding to the two parameters can be as shown in the following table (1), but not limited thereto.
Watch (1)
The strength degree beta of the social relationship among the users is calculated according to the following formula:
β=(γy+γm) (1)
and distinguishing whether the friend relationship exists between the accounts or not according to a preset rule and by combining with the value of the beta, wherein the preset rule can be as follows:
if the total number of the accounts with the group relationship is m, the number of the accounts with at least one friend relationship is n, and the ratio of the number of the accounts with the friend relationship to the total number of the accounts in the group is as follows:
in one embodiment, referring to fig. 3, the step S4 is followed by the step S5:
for the transaction determined as the swizzle transaction in the step S4, a further determination is made as to whether it is a swizzle transaction through a decision tree based on common evaluation features.
The decision tree based on the general evaluation feature, which is an evaluation feature of not all users and from which past evaluation data of each user can be used as a data source, further determines whether the transaction is a transaction swiped in step S4.
Preferably, the step of constructing the decision tree based on the general evaluation features comprises:
a1: setting a judgment standard of the bill-swiping transaction, and judging the evaluation behavior with the following characteristics as a group bill-swiping transaction: the evaluation rate exceeds a first preset threshold, the good evaluation rate exceeds a second preset threshold, and the first time evaluation rate exceeds a third preset threshold; the evaluation rate is high, the evaluation is many and good and the evaluation can be set in time, and the above characteristics can be determined by dividing through the set rules.
A2: extracting attribute features from multiple brush transaction and generating a training data set, wherein the attribute features comprise an evaluation rate, a goodness evaluation rate and a first-time evaluation rate; the evaluation rate is a ratio of the number of evaluation strokes to the number of transaction strokes, the evaluation rate is a ratio of the number of strokes giving good evaluation to the total number of evaluation strokes, and the first time evaluation rate is a ratio of the number of strokes giving evaluation to the total number of evaluation strokes at the first time, and the first time can be set as needed.
A3: and B, calculating the information gain of each attribute feature contained in the training data set by combining the judgment standard of the step A1, and selecting a split node according to the information gain to construct the decision tree based on the general evaluation feature. The information gain can be used for quantifying the attribute characteristics to a certain extent, and the information gain can be calculated through informatics knowledge, which is not described herein again. The decision tree formed by the embodiment of the invention is shown in fig. 2.
Preferably, step S5 includes:
s51: for the transaction determined as the transaction of the order-swiped transaction in the step S4, acquiring corresponding commodity evaluation information including an evaluation rate, evaluation quality and evaluation timeliness;
s52: and calculating the price rate, the good evaluation rate and the first time evaluation rate according to the commodity evaluation information, and judging whether the transaction is a bill-swiping transaction or not through the decision tree based on the general evaluation features according to the calculation result.
And judging whether the transaction is a one-way-swiping behavior or a social group-spreading purchasing behavior (namely, the one-way-swiping behavior is not) according to the obtained commodity evaluation information related to the interactive account in a certain commodity of the merchant, including the evaluation rate, the evaluation quality and the evaluation timeliness degree.
The invention is further explained below with reference to a specific embodiment:
the obtained historical social information corresponding to 10 accounts on the YY voice platform is as follows (2):
watch (2)
And binding the equipment and equipment identification codes (namely the equipment and equipment identification codes in the same row in the table) logged in the same YY voice account in the table (2) with each other and with the YY voice account.
Acquiring historical transaction information of 100 users on the Taobao, and binding the equipment and the equipment identification code of the 100 users logged in the Taobao account with each other and the Taobao account, wherein 10 users are taken as an example for explanation, and the historical interaction information of the 10 users is as shown in the following table (3):
watch (3)
And (3) binding the equipment and the equipment identification codes (namely the equipment and the equipment identification codes in the same row in the table) logged in the same panning account with each other and binding the equipment and the equipment identification codes with the corresponding panning account.
Acquiring 10 transaction records of a certain shop on a Taobao with a certain pair of shoes in one day, wherein the detailed information is as shown in the following table (4):
watch (4)
Firstly, for a first transaction: the using equipment is a tablet computer, and the MAC address is 3C: d0: f8: 2C: 99: a2, the device identification code bound with the MAC address in the lookup table (3) is a mobile phone number 13205812397 and a mobile phone IMEI code 354717046702222, and the device identification code bound with the MAC address in the lookup table (2) is searched with the code of 3C: d0: f8: 2C: 99: the identical or bound device identification codes A2 and/or 13205812397 and/or 354717046702222 find the device identification code with the row number 2 in the table (2), which comprises the IMEI code of the mobile phone: 354717046702222 and tablet MAC address: 3C: d0: f8: 2C: 99: a2 with a YY voice account number oflisi@qq.com。
For the second transaction, the used equipment is a computer, the MAC address of the computer is 00-01-6C-5D-46-93, and the equipment identification code bound with the MAC address in the lookup table (3) is a mobile phone number: 13100002345 and handset IMEI code: 866723019683291, searching the device identification code which is the same as or bound to 00-01-6C-5D-46-93 and/or 13100002345 and/or 866723019683291 in the table (2), and finding the device identification code of the row with the serial number of 1 in the table (2), wherein the device identification code comprises the mobile phone number: 13100002345, IMEI code 866723019683291 of mobile phone and MAC address 00-01-6C-5D-46-93 of computer, and the YY voice account number iszhangsan@qq.com。
For the third transaction, the used equipment is a computer, the MAC address of the computer is 00-0C-29-58-99-11, no other equipment identification code is bound with the address in the lookup table (3), the equipment identification code which is the same as or bound with the 00-0C-29-58-99-11 is searched in the table (2), and the equipment identification code of the row with the serial number of 4 in the table (2) is found and comprises the MAC address of the computer: 00-0C-29-58-99-11 and tablet computer MAC address 4E: 33: 4B: 33: AB: 6H with a YY voice account ofxiaohong@126.com。
And fourthly, for the fourth transaction, the used equipment is a mobile phone, and the corresponding equipment identification code comprises a mobile phone number: 15821052670 and handset IMEI code: 860859027401111, the equipment identification code bound with the mobile phone number and the mobile phone IMEI code in the lookup table (3) is 00-19-21-C4-BA-DB, and the equipment identification code bound with the mobile phone number in the lookup table (2) is as follows: 15821052670 and handset IMEI code: 860859027401111 and MAC address 00-19-21-C4-BA-DB, finding out the equipment identification code with serial number 10 in the row, including IMEI code 860859027401111 of mobile phone and MAC address 00-19-21-C4-BA-DB, and the corresponding YY voice account number isAbc1256@126.com。
For the fifth transaction, the used equipment is a tablet computer, and the MAC address is C8: 33: 4B: 33: AB: d7, the device identification code bound with the MAC address in the lookup table (3) is a mobile phone number 13958013467 and a mobile phone IMEI code 012425003304271, and the device identification code bound with the MAC address in the lookup table (2) is similar to the device identification code C8: 33: 4B: 33: AB: d7 and/or 012425003304271 are identical or bound to find the line with the serial number of 3 in the table (2), and the corresponding YY voice account number iswangwu@163.com。
For the sixth transaction, the used equipment is the mobile phone, the IMEI code of the mobile phone is 863664000059021, the equipment identification code bound with the mobile phone in the lookup table (3) is the MAC address 5B of the tablet computer: 3D: 11: CD: 5E: 6C. And 863664000059021 and/or 5B in the lookup table (2): 3D: 11: CD: 5E: 6C the same or bound equipment identification code, find the row with the serial number of 6 in the table (2), and the corresponding YY voice account number isxiaoming@qq.com。
Seventhly, for the 7 th, 8 th, 9 th and 10 th transactions, the YY voice account corresponding to the 7 th transaction is 23657890, and the YY voice account corresponding to the 8 th transaction is 23657890xiaohua@163.comThe YY voice account corresponding to the 9 th transaction is 78459944, and the YY voice account corresponding to the 10 th transaction is 654987321.
And (3) according to the YY voice account found in the step, finding the group relation of each YY voice account by combining the group information of the YY voice account in the table (2).
Taking intersection of the groups in the table (2) to find outzhangsan@qq.com、wangwu@163.com、xiaohong@ 126.com、23657890、xiaoming@qq.com、78459944、xiaohua@163.comIf the 7 YY voice account numbers have a group relationship, the 7 transactions of the 1 st, 3 rd, 4 th, 5 th, 6 th, 7 th and 9 th are preliminarily determined as the bill swiping.
Considering whether a friend relationship exists between the 7 YY voice accounts with the group relationship, in the embodiment of the present invention, whether a friend relationship exists between users of the mobile phone can be comprehensively determined according to the information of the contact between the mobile phone numbers and the information of the contact between the YY voice accounts, where the specific information is as follows (5):
watch (5)
Based on the above table (5) and the formula of calculating β, the degree of strength of social association between users is calculated:
βAB=0.2;βBC=1.3;βCD=0.2;βDE=0.1;βEF=0.4;βFG=1.3;βAG0.2 … … wherein, βABIs the social strength of the association between user A and user B, betaACThe strength of the social association between user a and user C, and so on.
And according to
And judging that friends are arranged between B and C and friends are arranged between F and G, and calculating the relationship of the friends in sequence. Then calculating the ratio of the account number with friend relationship to the total account number in the group
For example, if n is 8 and m is 10, p is 0.8.
Thereby determining that the transaction is a swiped transaction.
And further judging whether the transaction judged as the bill-swiping in the steps is the bill-swiping or not through a decision tree based on the general evaluation features, wherein the specific steps are as follows: acquiring the transaction judged to be the bill-swiped, and if the 5 transactions judged to be the suspected bill-swiped in the above step are acquired, acquiring the evaluation record information of the 5 transactions, as shown in the following table (6):
watch (6)
By setting specific evaluation rate threshold values such as 0.8, good evaluation rate threshold value 0.9 and first time evaluation rate threshold value 0.5, the first time can be set within 3 hours, because the brush is generally very timely to give evaluation. And (4) once more, the decision tree is used according to the acquired information of the table (6), and then the 5 transactions judged to be suspected to be the transactions can be judged to be the last transactions for brushing or not.
The invention also provides a computer device, which comprises a memory, a processor and a computer program stored on the memory and capable of being called by the processor, wherein when the processor executes the computer program, the method for detecting a swipe bill based on the device identification code and the social group information is implemented as described in any one of the foregoing embodiments.
The present invention also provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method for detecting a swipe form based on a device identifier and social group information as described in any one of the foregoing embodiments.
The invention also provides a system for detecting the bill swiping based on the equipment identification code and the social group information, which comprises the following steps:
the information acquisition module is used for acquiring social group information on the social platform and interaction information of consumers and commodities in a certain time period on the e-commerce platform; the social group information comprises a social account number, a group where the social account number is located and an equipment identification code used when the social account number is in social connection; the interactive information comprises an interactive account, an interactive commodity identifier and a device identifier used when the interactive account is interacted;
the account matching module is used for executing association between the equipment identification code used when the social contact occurs to the social account and the equipment identification code used when the interactive account interacts, and matching each social account with the interactive account;
the bill-swiping initial judgment module is used for acquiring a plurality of transaction records to be detected of a commodity, selecting a plurality of social account numbers corresponding to interactive account numbers with transactions, judging whether a group relationship exists or not, and if the groups of the social account numbers are the same, judging the corresponding transactions as bill-swiping transactions.
According to an embodiment of the present invention, the system for detecting a swipe form based on the device identification code and the social group information further includes:
and the bill-swiping secondary judgment module executes the transaction which is judged as the bill-swiping transaction in the bill-swiping primary judgment module, detects whether a friend relationship exists between the corresponding social account numbers with the group relationship, judges that the bill-swiping transaction is not the bill-swiping transaction if the social account numbers with the friend relationship in the group reach a preset ratio, and judges that the bill-swiping transaction is the bill-swiping transaction if the social account numbers with the friend relationship in the group do not reach the preset ratio.
According to an embodiment of the present invention, the system for detecting a swipe form based on the device identification code and the social group information further includes:
and the list-swiping three-judgment module is used for further judging whether the transaction is the list-swiping transaction or not through a decision tree based on the general evaluation features aiming at the transaction which is judged to be the list-swiping transaction in the list-swiping two-judgment module.
For specific contents of the system for detecting a swipe bill based on the device identifier and the social group information according to the present invention, reference may be made to the description of the method for detecting a swipe bill based on the device identifier and the social group information in the foregoing embodiments, and details are not repeated here.
According to the embodiment of the invention, the transaction information and the social information can be obtained, the login record information of the computer-side software (WEB page and the like) or the mobile phone-side software at the same/different time of different equipment is matched, the corresponding relation between the interactive account and the social account is found by using the equipment identification code of the equipment, and whether the group relation exists between the social accounts corresponding to the interactive account is judged, so that a great amount of row-refreshing actions realized by the internal organization and the association mode between the social software are detected.
The embodiment of the invention can effectively overcome the defect that the traditional identification method for the false transaction screening system adopting the traditional list-swiping characteristic can not screen the high-fidelity characteristic list-swiping activity of the false transaction industrial chain adopting the real logistics number, the real transaction behavior and the information of the real purchaser and the addressee, and can effectively detect the organized list-swiping behavior of the list-swiping organization through the social platform software.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the scope of the claims, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention.