CN110009372B - User risk identification method and device - Google Patents
User risk identification method and device Download PDFInfo
- Publication number
- CN110009372B CN110009372B CN201910114022.3A CN201910114022A CN110009372B CN 110009372 B CN110009372 B CN 110009372B CN 201910114022 A CN201910114022 A CN 201910114022A CN 110009372 B CN110009372 B CN 110009372B
- Authority
- CN
- China
- Prior art keywords
- user
- users
- seller
- risk
- buyers
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The application provides a user risk identification method and a user risk identification device, which can distinguish the severity of behaviors of different users and process the user according to the severity of the behaviors. The method comprises the following steps: acquiring behavior information of each user, and distinguishing various associated users of each user according to the behavior information, wherein the associated users are users interacting with the behavior information of the users; for each user, weighting each type of associated user of the user according to the respective preset risk weight to obtain a risk score of the user; and according to the risk scores of the users, performing corresponding monitoring and limiting operations on the users.
Description
Technical Field
The present application relates to data processing technology, and in particular, to a method and apparatus for identifying risk of a user.
Background
With the popularity of computers and the development of networks, networks are increasingly important in life. The user can transfer information on the social network platform and can trade shopping in the electronic trading platform.
However, the information on the network is not all secure and reliable. For example, on a social networking platform, some users may disseminate rumors, and some users may attack others. As another example, in electronic trading platforms, some users may conduct fraudulent transactions in order to increase confidence, and some users may sell counterfeit and inferior products.
The data of each day running and interacting in the platform is hundreds of millions, and the data related to the above behaviors of the users are very hidden, so that the manager of the platform can find and find the root of the problem only when the behaviors are greatly influenced or reported by other users. And then find the user who performs these actions and penalize, for example, perform the measures of checking up the account number of the user and adding the user to a blacklist, the user can not log on the corresponding platform by using the account number any more.
However, the above-mentioned actions of users are slightly and severely different, for example, in the electronic commerce platform, some users intentionally sell counterfeit products to make a profit as genuine products, and some users just make a false transaction when new products are on the market, only for improving the confidence of sellers, and the products themselves have no problem.
Therefore, one technical problem that is urgent to be solved by those skilled in the art is to propose a user risk recognition method, which can distinguish the severity of behaviors of different users, and perform distinguishing treatment on the users according to the severity of the behaviors.
Disclosure of Invention
The application provides a user risk identification method and a user risk identification device, which can distinguish the severity of behaviors of different users and perform distinguishing treatment on the users according to the severity of the behaviors.
In order to solve the above problems, the present application discloses a user risk identification method, comprising:
acquiring user behavior information of a seller, and determining at least one type of associated users of the seller according to the behavior information, wherein the associated users of the seller are users performing interaction behaviors with the seller;
calculating the risk score of the seller according to the determined number of various associated users and the preset risk weights of various associated users, wherein the risk score is used for evaluating the risk severity of the seller;
wherein the associated user of the seller comprises at least one of: shadow users, professional bad evaluation users, bulk buyers, fan users and letter frying buyers; the determining at least one type of associated user of the seller according to the behavior information comprises at least one of the following steps:
Determining a user having at least one of the following relationships with a seller as a shadow user of the seller: the same harvest address is adopted, the same IP address is adopted for logging in, and the same contact mode is adopted;
collecting transaction evaluation information of a plurality of sellers, acquiring buyers rated by a first seller, and determining professional rated users employed by the buyers for a second seller if the money purchased by the buyers comes from the second seller or the goods purchased by the buyers are mailed to the registration address of the second seller;
counting a plurality of buyers of the seller, obtaining transaction amount of each buyer and the seller, ranking transaction amount weights of each buyer, and determining the buyers ranked at the top as bulk buyers of the seller;
counting a plurality of buyers of a seller, acquiring transaction amounts of each buyer and the seller, and determining the buyers with transaction amount weights exceeding a threshold as fan users of the seller;
if the price of the commodity sold to the buyer by the seller is lower than the commodity price, determining that the buyer is a letter-frying buyer of the seller; and/or if the money transfer account numbers used by multiple buyers of the seller are the same, determining that the multiple buyers are letter-fried buyers of the seller.
The application also provides a user risk identification device, which comprises:
the acquisition and distinguishing module is used for determining at least one type of associated users of the seller according to the behavior information, wherein the associated users of the seller are users performing interaction behaviors with the seller;
the risk score acquisition module is used for calculating the risk score of the seller according to the determined number of various associated users and the preset risk weights of the various associated users, and the risk score is used for evaluating the risk severity of the seller;
wherein the associated user of the seller comprises at least one of: shadow users, professional bad evaluation users, bulk buyers, fan users and letter frying buyers; the determining at least one type of associated user of the seller according to the behavior information comprises at least one of the following steps:
determining a user having at least one of the following relationships with a seller as a shadow user of the seller: the same harvest address is adopted, the same IP address is adopted for logging in, and the same contact mode is adopted;
collecting transaction evaluation information of a plurality of sellers, acquiring buyers rated by a first seller, and determining professional rated users employed by the buyers for a second seller if the money purchased by the buyers comes from the second seller or the goods purchased by the buyers are mailed to the registration address of the second seller;
Counting a plurality of buyers of the seller, obtaining transaction amount of each buyer and the seller, ranking transaction amount weights of each buyer, and determining the buyers ranked at the top as bulk buyers of the seller;
counting a plurality of buyers of a seller, acquiring transaction amounts of each buyer and the seller, and determining the buyers with transaction amount weights exceeding a threshold as fan users of the seller;
if the price of the commodity sold to the buyer by the seller is lower than the commodity price, determining that the buyer is a letter-frying buyer of the seller; and/or if the money transfer account numbers used by multiple buyers of the seller are the same, determining that the multiple buyers are letter-fried buyers of the seller.
Compared with the prior art, the application at least comprises the following advantages:
first, the impact of the user's behavior is slightly and severely different from that of the prior art. The application acquires the behavior information of each user, and distinguishes various associated users of each user according to the behavior information, and different associated users can have different effects on the behavior of the user. As in the e-commerce platform, the act of the seller hiring the letter buyer for a false transaction is relatively slight, while the act of the seller hiring the professional bad evaluator for improper competition is relatively serious. Therefore, risk weights are preset for each associated user, and each type of associated user of the user can be weighted according to the respective preset risk weights for each user, so that the risk scores of the users are obtained. And evaluating the severity of the behavior of the user through the risk score, and further carrying out corresponding monitoring and limiting operation on the user according to the risk score of the user.
And secondly, the method can also take the penalized user as the first garbage user, presets the first score of the first garbage user, and adds the first score into the risk score corresponding to the first garbage user, so that the influence of the history of the user on the risk score is considered, and the risk score is more accurate and comprehensive.
And the application can be applied to the field of electronic commerce, and the seller is taken as the user of the application, so that the associated user corresponding to the seller comprises at least one of the following: shadow users, professional bad evaluation users, bulk buyers, fan users and letter-frying buyers. Through analysis and identification of the associated users, it is possible to obtain whether each seller is engaged in a malignant competition or a spurious transaction. The risk score can be used as one of credit evaluation standards of the seller in electronic commerce and can also be related to a corresponding natural person, and the application field is very wide as one of credit evaluation standards in society.
Drawings
FIG. 1 is a flowchart of a user risk identification method according to an embodiment of the present application;
FIG. 2 is a flowchart of a second garbage user identification process in a user identification method according to a preferred embodiment of the present application;
FIG. 3 is a flowchart of a third garbage user identification process in a user identification method according to a preferred embodiment of the present application;
fig. 4 is a block diagram of a user risk identification device according to an embodiment of the present application.
Detailed Description
In order that the above-recited objects, features and advantages of the present application will become more readily apparent, a more particular description of the application will be rendered by reference to the appended drawings and appended detailed description.
The effects caused by the behavior of the user are slightly and severely different, which cannot be distinguished by the prior art.
The application provides a user risk identification method which can distinguish the severity of behaviors of different users and process the user according to the severity of the behaviors.
The process according to the application is described in more detail by way of examples.
Referring to fig. 1, a flowchart of a user risk identification method according to an embodiment of the present application is provided.
Step 11, obtaining behavior information of each user, and distinguishing various associated users of each user according to the behavior information, wherein the associated users are users interacting with the behavior information of the users;
the user performs network activities such as shopping, publishing microblogs and the like in the platform, and corresponding behavior information can be generated. The application can obtain the user ID (identity) of each user in the platform, then inquire all message records of the user ID in the platform database, obtain the behavior information of the user from the message records, and analyze the behavior information. The behavior information can comprise a plurality of categories, various associated users of each user can be distinguished through the behavior information of different categories, and corresponding user IDs are obtained, for example, various associated users of one user in the microblog comprise vermicelli, mutual meal and the like, and various associated users of one seller comprise bulk buyers, letter-frying buyers and the like in an electronic commerce platform.
The associated user is a user interacting with the behavior information of the user. For example, an associated user at user 1 in a microblog may include: a user who user 1 focuses on, a user who focuses on user 1, a user who user 1 focuses on each other and a user who does not focus on user 1 but forwards a microblog, and the like. In an e-commerce web site, an associated user of a seller may include a buyer.
Step 12, weighting each type of associated user of the user according to the respective preset risk weight for each user to obtain a risk score of the user;
a user interacting with an associated user may indicate that the user is at risk to some extent, e.g., an associated user of a seller includes a letter of frying buyer, the seller may have a false transaction, and e.g., an associated user of a seller includes a professional bad evaluator, the seller may have performed undue competition. The risk degrees of various associated users are different, so that the risk weights of various associated users are preset, the weights can be set according to different platforms in actual treatment, and the risk degree of the various associated users is not limited by the risk degree setting method.
And if the associated users of different categories have respective risk weights, analyzing the risk scores of the users after analyzing the various associated users of each user, namely, calculating results after weighting the various associated users of the users according to the respective preset risk weights.
For example, two kinds of risk users of the user X are detected, each including 10 first kind of risk users and 3 second kind of risk users, the first kind of risk users has a weight of 0.3, and the second kind of risk users has a weight of 0.5, and the risk score of the user X is 10×0.3+3×0.5=4.5
And 13, carrying out corresponding monitoring and limiting operation on the user according to the risk score of the user.
Performing monitoring and limiting operations corresponding to the risk scores of the users according to the risk scores of the users,
for example, when a user applies for loan in an e-commerce platform, all users 1-3 apply for loan for 10 ten thousand yuan, and it is detected that the risk score of user 1 is 60, the risk score of user 2 is 20, and the risk score of user 3 is 85, then 5 ten thousand yuan can be given to user 1, 10 ten thousand yuan can be given to user 2, and no loan is given to user 3.
In summary, the influence caused by the behavior of the user is slightly and severely different, and the prior art cannot distinguish the influence. The application acquires the behavior information of each user, and distinguishes various associated users of each user according to the behavior information, and different associated users can have different effects on the behavior of the user. As in the e-commerce platform, the act of the seller hiring the letter buyer for a false transaction is relatively slight, while the act of the seller hiring the professional bad evaluator for improper competition is relatively serious. Therefore, risk weights are preset for each associated user, and each type of associated user of the user can be weighted according to the respective preset risk weights for each user, so that the risk scores of the users are obtained. And evaluating the severity of the behavior of the user through the risk score, and further carrying out corresponding monitoring and limiting operation on the user according to the risk score of the user.
Preferably, the method may further comprise:
according to the history record, the penalized user is used as a first junk user, and behavior information of the first junk user is obtained; and adding a first score of the preset first junk user into the risk score.
If a user is engaged in the propagation of bad information in a platform, the user is penalized by an administrator of the platform after serious behavior is found by the administrator. For example, on a social network platform, one user attacks other users, the users can be prohibited from commenting on the platform, and the account can be sealed when the scenario is serious. As another example, in an e-commerce platform, a user may sell counterfeit products and may be penalized. The platform will record all penalties and store them in a database.
Therefore, the method and the system can also acquire the penalized users from the database, and take the penalized users as the first garbage users, for example, all penalized users in the platform can be recorded in the database, and correspondingly the penalty records can be searched in the database, wherein all the user Identifications (IDs) of the penalized users are recorded in the penalty records, and the users identified by the user IDs are the first garbage users. And then searching a message record corresponding to the user ID, and acquiring the behavior information of the first junk user from the message record. In addition, a certain risk exists in the transaction with the first junk user, the first score of the junk user is preset in the method, and when the user is identified, if the user is detected to be the first junk user, the corresponding first score is added into the risk score of the user.
For example, a first score of 20 is preset, and if it is detected that the user X is penalized in the above example, the user X is the first garbage user, and the risk score of the user X is 4.5+20=24.5.
In summary, the method and the system can also take the penalized user as the first garbage user, preset the first score of the first garbage user, and add the first score into the risk score corresponding to the first garbage user, so that the influence of the history of the user on the risk score is considered, and the risk score is more accurate and comprehensive.
The application can be applied to the field of electronic commerce, in an electronic commerce platform, a seller is taken as a user of the application, and the associated user corresponding to the seller comprises at least one of the following: shadow users, professional bad evaluation users, bulk buyers, fan users, letter-fried buyers and risk users.
When acquiring behavior information of a seller, each transaction can be acquired by taking the transaction as a unit: seller, buyer, commodity name, category, price, purchase time, payment time, shipping time, receiving time, transaction evaluation content, etc.
And then taking the seller as a unit, acquiring the purchasing amount, the time preference of the buyer group of the seller, the transaction trend and the evaluation related condition.
And taking the buyer as a unit to acquire the main consumption type, the amount and the distribution of sellers of the buyer.
Therefore, in the implementation of the application, the relationship diagram of each seller can be built by centering on the seller, so that the relationship of each seller, such as shadow users, professional bad evaluation users, bulk buyers, fan users, letter-fried buyers, risk users and the like, can be clearly identified.
Preferably, the distinguishing each type of associated user of each user according to the behavior information includes at least one of the following:
1) Analyzing login information, logistics information, registration information and contact information in behavior information of a seller to acquire a shadow user of the seller;
in an e-commerce platform, a natural person user is usually allowed to register only one account, i.e. a natural person only obtains one user ID, but in actual processing, there are often some buyers and sellers that use different identity information to register multiple accounts for improper interests, and these accounts are called shadow accounts, i.e. a user has multiple user IDs. For example, a natural person registers 3 accounts, namely accounts 1-3, with 3 mailboxes of the natural person, and for account 1, accounts 2 and 3 are shadow accounts of account 1, and the user of the account is a shadow user, and the user ID corresponding to the account is the user ID of the shadow user.
Therefore, the behavior information of the seller can be analyzed, and login information, logistics information, registration information, contact information and the like in the behavior information can be analyzed, for example, it is found that the same receiving address is adopted by a plurality of accounts, and for example, the same IP address is often adopted by a plurality of accounts to log in, and the same contact manner is also possibly adopted by a plurality of accounts. Therefore, through analysis of the information, whether the seller has a shadow account number can be obtained, and if the seller has the shadow account number, the user ID corresponding to each shadow account number is the user ID of the shadow user of the seller.
2) Collecting and analyzing transaction evaluation information in the behavior information of all sellers to obtain professional evaluation users, and analyzing sellers interacting with the professional evaluation users;
the job review user refers to a buyer who purchases goods or services of an e-commerce platform seller designated by a competitor (a seller), and maliciously defaults the purchased goods or services after the purchase is finished, so that the seller can strike the competitor, and the buyer, i.e. the job review user, can make a profit from the buyer.
The application can collect and analyze transaction evaluation information in the behavior information of all sellers to obtain professional bad evaluation users, for example, some buyers frequently give the sellers bad evaluation, obtain the user IDs of the buyers, namely the buyers need to be subjected to important monitoring, and the analysis of the behavior information corresponding to the user IDs of the buyers finds that the money purchased by the buyers comes from the seller of another trade, possibly the commodity purchased by the buyers is mailed to the registration address of the seller of another trade, and the buyers are the professional bad evaluation users, and the sellers corresponding to the seller providing the money and the mailing address interact with the professional bad evaluation users, namely the sellers hiring the professional bad evaluation users. Therefore, by analyzing the users frequently giving bad comments to the sellers, whether the users are bad-comment users or not can be analyzed, and if the users are bad-comment users, sellers interacting with the bad-comment users can also be analyzed.
3) Acquiring bulk buyers and fan users of sellers through weight information of transaction information in behavior information of sellers;
(1) the bulk buyer is the top n buyers in the transaction amount of a seller. For example, all buyers of the seller 01 are counted, transaction amounts of each buyer and the seller 01 are obtained, all buyers are ranked according to the transaction amounts, the result after the ranking is that the buyer A, the buyer D, the buyer C, the buyer E and the buyer B are the buyers A and the buyer D, and if the buyer ranked in the first 2 is the bulk buyer, the bulk buyer of the seller 01 is the buyer A and the buyer D. The buyers can be ranked according to the proportion (namely weight) occupied by the transaction amount of the transaction information in the behavior information of the sellers, and the bulk buyers of the sellers can be obtained.
(2) When the ratio of the transaction amount of a buyer to a specific seller and all the transaction amounts of the buyer exceeds a threshold, i.e. the weight of the transaction amount exceeds the threshold, the buyer is referred to as the fan user of the specific seller. For example, 90% of all transactions by buyer D are purchased by seller 01, i.e., seller 01 has a weight of 0.9 for buyer D, and if the threshold is 0.75, buyer D is the fan user of seller 01.
4) And analyzing the buyer information, the logistics information, the payment information and the evaluation information in the behavior information of the seller to obtain the stir-frying buyer of the seller.
The stir-fried buyers are buyers participating in false online transactions, improving the credit level of shops and manufacturing goods or serving hot sales. For example, if seller 01 employed buyer C for a dummy transaction to promote the credit rating of the store, and the bid price of the merchandise was 1000, then the price that seller 01 sold to buyer C might be 100. For another example, it is found that in the payment information of seller 01, there are 5 different accounts, i.e., the remittance sources of different user IDs are the same account. Therefore, by analyzing the behavior information of the seller, buyer information, logistics information, payment information, evaluation information and the like can be obtained from the behavior information
Referring to fig. 2, a second garbage user identification flowchart in a user identification method according to a preferred embodiment of the present application is shown.
Preferably, in the electronic commerce platform, after taking a certain penalized seller as the first garbage user, the method further comprises:
step 21, analyzing all the associated users of the first junk user through the behavior information of the first junk user, and calculating the weight of each associated user of the first junk user;
After the related user, namely the buyer, acquires the user ID of the first junk user, and then acquires the behavior information of the first junk user through the user ID, further analyzes the behavior information of the first junk user, for example, searches the related information interacted with the behavior information, such as the information for evaluating the behavior information, by taking the behavior information of the first junk user as a source. And searching the associated user which issues the associated information through the associated information, namely analyzing all buyers of the first garbage user, and simultaneously calculating the weight of each buyer of the first garbage user.
For example, the number of commodities purchased by the buyer and the total price of the commodities purchased by the buyer that transact with the first garbage user may be acquired, and then the ratio of the number of commodities purchased by each buyer to the total price of the commodities purchased by the buyer is calculated, and the ratio is taken as the weight of the buyer in the garbage user.
And step 22, analyzing a first junk user corresponding to the associated user by taking any one of the analyzed associated users as a dimension, and identifying the associated user as a risk user according to the first junk user number corresponding to the associated user and the weight of the associated user in the corresponding first junk user.
And taking the associated user as a dimension, namely taking the buyer as a dimension, for example, taking the user ID of the associated user as an identifier, acquiring the user ID of the bad user corresponding to the identifier, namely acquiring the first junk users corresponding to each buyer, and then identifying the associated user as a risk user according to the number of the first junk users corresponding to the associated user and the weight of the associated user in the corresponding first junk users.
For example, if the number of first junk users corresponding to the buyers is detected to exceed the number threshold, the buyers are risk users. For another example, if the number of first garbage users corresponding to the buyer does not exceed the number threshold, but the weight of the buyer is abnormal in the corresponding first garbage users, if the first garbage users are selling computers, the average price is more than 3000, but the price of the first garbage users sold to the buyer 001 is 100, the weight of the buyer 001 is abnormal in the first garbage users, and the buyer 001 is a risk user.
Step 23, obtaining behavior data of the risk user through clustering behavior information of the risk user, and taking the user with the behavior data as other risk users;
after the risk users are identified, the categories of some common behaviors of the risk users can be obtained through clustering by clustering the behavior information of the risk users and used as the behavior data of the risk users, namely, some common behavior modes of the risk users are obtained. And then taking the users with the clustered common behaviors, namely the users with the behavior data, as other risk users.
And step 24, searching a second junk user through other risk users, and adding a second score of the preset second junk user into the risk score.
And searching a second junk user through other risk users, for example, searching a seller which carries out transactions through the user IDs of the other risk users, calculating the weight and other data of the user IDs of the other risk users in corresponding sellers, if the weight is abnormal, taking the seller as the second junk user, namely, the user with the similar behavior of the first junk user, presetting a second score of the second junk user, and adding the second score into the risk score of the second junk user.
Referring to fig. 3, a third garbage user identification flowchart in a user identification method according to a preferred embodiment of the present application is provided.
Preferably, in the electronic commerce platform, the penalized seller is taken as a first garbage user, and the method further comprises:
step 31, clustering behavior information of all first junk users to obtain various first behavior data of the first junk users;
and acquiring behavior information of the first junk user, and clustering the penalized sellers by a commodity ID, commodity price, user ID of a buyer, a buyer nickname, registration time, registration source, receiving address, evaluation score, evaluation content, evaluation time, seller nickname, registration time, seller grade and several dimensions of the buyer grade in the behavior information to acquire various first behavior data of the first junk user.
Wherein, the behavior information may be as shown in table 1:
table 1 is of course only one form of behavioral information and is presented herein for exemplary discussion and should not be construed as limiting the application.
And then clustering the behavior information of all the first garbage users in the platform, for example, by a clustering method such as K-Means, and the like, so as to grasp the indexes which are mutually independent and have crowd distinction, namely the first behavior data. For example, transactions with 70% or more of the first type of first behavioral data are performed after mid-night, and transactions with 90% or more of the second type of first behavioral data are favored.
Step 32, randomly acquiring a plurality of users and behavior information of the users, clustering the behavior information of the users, and acquiring various second behavior data of the users;
in order to ensure the accuracy and stability of the clustering result of the behavior information of the first junk user, namely to ensure that the first behavior data is only the common behavior of the first junk user, but not the common behavior of all users. On the selection of the samples, attention is paid to the selection of random samples, namely, all sellers randomly acquire a plurality of users and behavior information of the users according to natural distribution, and perform clustering for a plurality of times to acquire various second behavior data of the users.
For example, transactions with a class two behavioral data of 90% or more are favored.
Step 33, comparing the first behavior data of the first junk user with the second behavior data of the user to determine first behavior data unique to the first junk user;
and then distinguishing whether the first behavior data is the common behavior of the first junk user or the common behavior of all users, and comparing the first behavior data of the first junk user with the second behavior data of the user to further determine the first behavior data unique to the first junk user. More than 70% of the transactions for the first behavioral data unique to the example above are performed after mid-night.
The method enables the finally generated clustering result to have obvious effects on the screening of the crowd.
And step 34, taking the user with the unique first behavior data as a third garbage user, and adding a third score of the preset third garbage user into the risk score.
Then, the users with the unique first behavior data have common behaviors with the first garbage users in all sellers, so that the users with the unique first behavior data can be used as third garbage users, third scores of the third garbage users are preset, and the third scores are added into risk scores of the third garbage users.
After the risk score of the seller is obtained by the method, the seller can perform corresponding monitoring and limiting operations according to the risk score.
For example, the risk score may be used as one of credit evaluation criteria of the seller, and whether to loan the seller, the amount of the loan, and the like may be determined based on the risk score.
The electronic commerce platform is used as a seller to carry out a real-name system, so that the seller can be related to a natural person, the risk score is used as one of credit evaluation standards of the natural person, and the natural person can evaluate according to the risk score when carrying out loan and business activities, so that the application field is very wide.
In summary, the present application may be applied to the field of electronic commerce, and the seller is taken as the user according to the present application, and the associated user corresponding to the seller includes at least one of the following: shadow users, professional bad evaluation users, bulk buyers, fan users, letter-fried buyers and risk users. Through analysis and identification of the associated users, it is possible to obtain whether each seller is engaged in a malignant competition or a spurious transaction. The risk score can be used as one of credit evaluation standards of the seller in electronic commerce and can also be related to a corresponding natural person, and the application field is very wide as one of credit evaluation standards in society.
Referring to fig. 4, a structure diagram of a user risk identification device according to an embodiment of the present application is provided.
Correspondingly, the application also provides a user risk identification device, which comprises: an acquisition and differentiation module 11, a risk score acquisition module 12 and a monitoring module 13, wherein:
the acquiring and distinguishing module 11 is configured to acquire behavior information of each user, and distinguish various associated users of each user according to the behavior information, where the associated users are users that interact with the behavior information of the user;
the risk score obtaining module 12 is configured to weight, for each user, each type of associated user of the user according to a preset risk weight, to obtain a risk score of the user;
and the monitoring module 13 is used for carrying out corresponding monitoring and limiting operation on the user according to the risk score of the user.
Preferably, the device further comprises:
the garbage user acquisition module is used for acquiring the penalized user as a first garbage user and acquiring behavior information of the first garbage user;
and the first adding module is used for adding the preset first score of the first garbage user into the risk score.
Preferably, in the electronic commerce platform, the user is a seller, and the associated user includes at least one of the following: shadow users, professional bad evaluation users, bulk buyers, fan users, letter-fried buyers and risk users.
Preferably, the acquiring and differentiating module 11 comprises at least one of the following:
the shadow user acquisition module is used for analyzing login information, logistics information, registration information and contact information in the behavior information of the seller to acquire a shadow user of the seller;
the professional evaluation user analysis module is used for summarizing and analyzing transaction evaluation information in the behavior information of all sellers, acquiring professional evaluation users and analyzing sellers interacting with the professional evaluation users;
the buyer and vermicelli acquisition module is used for acquiring bulk buyers and vermicelli users of the seller through weight information of transaction information in behavior information of the seller;
and the information stir-frying buyer acquisition module is used for acquiring the information stir-frying buyer of the seller by analyzing the buyer information, the logistics information, the payment information and the evaluation information in the behavior information of the seller.
Preferably, in the electronic commerce platform, the penalized seller is taken as a first garbage user, and the device further comprises a risk user identification module, and further comprises:
The analysis sub-module is used for analyzing all the associated users of the first junk user through the behavior information of the first junk user, and calculating the weight of each associated user of the first junk user;
and the identification sub-module is used for analyzing the first junk users corresponding to the associated users by taking the associated users as dimensions, and identifying the associated users as risk users according to the first junk users corresponding to the associated users and the weights of the associated users in the corresponding first junk users.
Preferably, the device further comprises:
the risk user clustering module is used for acquiring the behavior data of the risk users through clustering the behavior information of the risk users, and taking the users with the behavior data as other risk users;
and the second adding module is used for searching the second garbage user through other risk users and adding a second score of the preset second garbage user into the risk score.
Preferably, in the electronic commerce platform, the penalized seller is taken as the first garbage user, the device further comprises a third garbage user determining module, and the device further comprises:
The first clustering sub-module is used for clustering the behavior information of all the first garbage users and acquiring various first behavior data of the first garbage users;
the second aggregation sub-module is used for randomly acquiring a plurality of users and behavior information of the users, clustering the behavior information of the users and acquiring various second behavior data of the users;
the comparison and determination submodule is used for comparing the first behavior data of the first junk user with the second behavior data of the user and determining the first behavior data unique to the first junk user;
and the determining and adding sub-module is used for taking the user with the first behavior data as a third garbage user and adding a third score of the preset third garbage user into the risk score.
For the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.
As will be readily appreciated by those skilled in the art: any combination of the above embodiments is possible, and thus is an embodiment of the present application, but the present specification is not limited by the text.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the application.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
The foregoing has described in detail the method and apparatus for identifying risk of user provided by the present application, and specific examples have been applied herein to illustrate the principles and embodiments of the present application, the above examples being only for aiding in understanding the method and core idea of the present application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.
Claims (10)
1. A user risk identification method, comprising:
acquiring user behavior information of a seller, and determining at least one type of associated users of the seller according to the behavior information, wherein the associated users of the seller are users performing interaction behaviors with the seller;
calculating the risk score of the seller according to the determined number of various associated users and the preset risk weights of various associated users, wherein the risk score is used for evaluating the risk severity of the seller;
wherein the associated user of the seller comprises at least one of: shadow users, professional bad evaluation users, bulk buyers, fan users and letter frying buyers; the determining at least one type of associated user of the seller according to the behavior information comprises at least one of the following steps:
Determining a user having at least one of the following relationships with a seller as a shadow user of the seller: the same harvest address is adopted, the same IP address is adopted for logging in, and the same contact mode is adopted;
collecting transaction evaluation information of a plurality of sellers, acquiring buyers rated by a first seller, and determining professional rated users employed by the buyers for a second seller if the money purchased by the buyers comes from the second seller or the goods purchased by the buyers are mailed to the registration address of the second seller;
counting a plurality of buyers of the seller, obtaining transaction amount of each buyer and the seller, ranking transaction amount weights of each buyer, and determining the buyers ranked at the top as bulk buyers of the seller;
counting a plurality of buyers of a seller, acquiring transaction amounts of each buyer and the seller, and determining the buyers with transaction amount weights exceeding a threshold as fan users of the seller;
if the price of the commodity sold to the buyer by the seller is lower than the commodity price, determining that the buyer is a letter-frying buyer of the seller; and/or if the money transfer account numbers used by multiple buyers of the seller are the same, determining that the multiple buyers are letter-fried buyers of the seller.
2. The method of claim 1, further comprising:
taking the penalized seller as a first garbage user;
and adding a preset first score corresponding to the first junk user into the risk score calculated according to the number and the risk weight of the associated users.
3. The method of claim 1, further comprising:
taking the penalized seller as a first garbage user;
analyzing all associated users of a first junk user through behavior information of the first junk user, and calculating weight of each associated user in the first junk user;
and analyzing a first junk user corresponding to any associated user by taking the associated user as a dimension, and identifying whether the associated user is a risk user or not according to the first junk user quantity corresponding to the associated user and the weight of the associated user in the corresponding first junk user.
4. A method according to claim 3, further comprising:
acquiring behavior data of the risk user through clustering behavior information of the risk user, and taking the user with the behavior data as other risk users;
searching a second junk user through other risk users;
And adding a preset second score corresponding to the second garbage user into the risk score.
5. The method of claim 1, further comprising:
taking the penalized seller as a first garbage user;
clustering behavior information of all first junk users to obtain various first behavior data of the first junk users, wherein the first behavior data are used for representing common behaviors of the first junk users, and each type of first behavior data corresponds to one behavior feature;
randomly acquiring behavior information of a plurality of users and the plurality of users, clustering the behavior information of the plurality of users, and acquiring various second behavior data of the plurality of users, wherein the second behavior data is used for representing common behaviors of all the users, and each type of second behavior data corresponds to one behavior feature;
comparing the first behavior data of the first junk user with the second behavior data of the plurality of users to determine first behavior data unique to the first junk user;
and taking the user with the unique first behavior data as a third junk user, and adding a preset third score corresponding to the third junk user into the risk score.
6. A user risk identification device comprising:
the acquisition and distinguishing module is used for determining at least one type of associated users of the seller according to the behavior information, wherein the associated users of the seller are users performing interaction behaviors with the seller;
the risk score acquisition module is used for calculating the risk score of the seller according to the determined number of various associated users and the preset risk weights of the various associated users, and the risk score is used for evaluating the risk severity of the seller;
wherein the associated user of the seller comprises at least one of: shadow users, professional bad evaluation users, bulk buyers, fan users and letter frying buyers; the determining at least one type of associated user of the seller according to the behavior information comprises at least one of the following steps:
determining a user having at least one of the following relationships with a seller as a shadow user of the seller: the same harvest address is adopted, the same IP address is adopted for logging in, and the same contact mode is adopted;
collecting transaction evaluation information of a plurality of sellers, acquiring buyers rated by a first seller, and determining professional rated users employed by the buyers for a second seller if the money purchased by the buyers comes from the second seller or the goods purchased by the buyers are mailed to the registration address of the second seller;
Counting a plurality of buyers of the seller, obtaining transaction amount of each buyer and the seller, ranking transaction amount weights of each buyer, and determining the buyers ranked at the top as bulk buyers of the seller;
counting a plurality of buyers of a seller, acquiring transaction amounts of each buyer and the seller, and determining the buyers with transaction amount weights exceeding a threshold as fan users of the seller;
if the price of the commodity sold to the buyer by the seller is lower than the commodity price, determining that the buyer is a letter-frying buyer of the seller; and/or if the money transfer account numbers used by multiple buyers of the seller are the same, determining that the multiple buyers are letter-fried buyers of the seller.
7. The apparatus of claim 6, further comprising:
the garbage user acquisition module is used for taking the penalized seller as a first garbage user;
the first adding module adds a preset first score corresponding to the first junk user into the risk scores calculated according to the number of the associated users and the risk weights.
8. The apparatus of claim 6, further comprising:
the garbage user acquisition module is used for taking the penalized seller as a first garbage user;
The analysis sub-module is used for analyzing all the associated users of the first junk user through the behavior information of the first junk user, and calculating the weight of each associated user in the first junk user;
and the identification sub-module is used for analyzing the first junk users corresponding to any associated user by taking the associated user as a dimension, and identifying whether the associated user is a risk user or not according to the first junk user quantity corresponding to the associated user and the weight of the associated user in the corresponding first junk user.
9. The apparatus of claim 8, further comprising:
the risk user clustering module is used for acquiring the behavior data of the risk users through clustering the behavior information of the risk users, and taking the users with the behavior data as other risk users;
the second adding module is used for searching a second junk user through other risk users; and adding a preset second score corresponding to the second garbage user into the risk score.
10. The apparatus of claim 6, further comprising:
the garbage user acquisition module is used for taking the penalized seller as a first garbage user;
The first clustering sub-module is used for clustering the behavior information of all the first garbage users to obtain various first behavior data of the first garbage users, wherein the first behavior data are used for representing the common behavior of the first garbage users, and each type of first behavior data corresponds to one behavior characteristic;
the second aggregation sub-module is used for randomly acquiring behavior information of a plurality of users and the plurality of users, clustering the behavior information of the plurality of users and acquiring various second behavior data of the plurality of users, wherein the second behavior data is used for representing common behaviors of all the users, and each type of second behavior data corresponds to one behavior feature;
the comparison and determination submodule is used for comparing the first behavior data of the first junk user with the second behavior data of the plurality of users and determining first behavior data unique to the first junk user;
and determining and adding a sub-module, taking the user with the unique first behavior data as a third garbage user, and adding a preset third score corresponding to the third garbage user into the risk score.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910114022.3A CN110009372B (en) | 2012-08-03 | 2012-08-03 | User risk identification method and device |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910114022.3A CN110009372B (en) | 2012-08-03 | 2012-08-03 | User risk identification method and device |
| CN201210275476.7A CN103577991B (en) | 2012-08-03 | 2012-08-03 | The recognition methods of user a kind of and device |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210275476.7A Division CN103577991B (en) | 2012-08-03 | 2012-08-03 | The recognition methods of user a kind of and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110009372A CN110009372A (en) | 2019-07-12 |
| CN110009372B true CN110009372B (en) | 2023-08-18 |
Family
ID=50049727
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210275476.7A Active CN103577991B (en) | 2012-08-03 | 2012-08-03 | The recognition methods of user a kind of and device |
| CN201910114022.3A Active CN110009372B (en) | 2012-08-03 | 2012-08-03 | User risk identification method and device |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210275476.7A Active CN103577991B (en) | 2012-08-03 | 2012-08-03 | The recognition methods of user a kind of and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN103577991B (en) |
Families Citing this family (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104933570A (en) * | 2014-03-20 | 2015-09-23 | 阿里巴巴集团控股有限公司 | User detection method and device |
| CN103955833B (en) * | 2014-03-31 | 2017-08-15 | 浙江工商大学 | Waterborne troops's indentity identifying method based on wash sale and social networks matrix analysis |
| CN105243252B (en) * | 2014-07-09 | 2018-08-07 | 阿里巴巴集团控股有限公司 | A kind of method and device of account risk assessment |
| CN105335855A (en) * | 2014-08-06 | 2016-02-17 | 阿里巴巴集团控股有限公司 | Transaction risk identification method and apparatus |
| CN105447028B (en) * | 2014-08-27 | 2019-06-28 | 阿里巴巴集团控股有限公司 | The method and device of identification feature account |
| CN105631681B (en) * | 2014-10-29 | 2019-06-21 | 阿里巴巴集团控股有限公司 | A kind of detection method and equipment of electronic certificate transmission account |
| CN106033575A (en) * | 2015-03-11 | 2016-10-19 | 阿里巴巴集团控股有限公司 | Risk account identification method and apparatus |
| CN106161372B (en) * | 2015-04-09 | 2019-05-31 | 阿里巴巴集团控股有限公司 | A kind of Risk Identification Method and device based on address matching |
| CN105138897B (en) * | 2015-08-24 | 2019-04-16 | 百度在线网络技术（北京）有限公司 | Determine the method and device of user right |
| CN106682906B (en) * | 2015-11-10 | 2021-03-19 | 创新先进技术有限公司 | Risk identification and service processing method and equipment |
| CN105307176B (en) * | 2015-11-10 | 2019-03-08 | 中国科学院信息工程研究所 | Robustness message routing method in a kind of mobile social opportunistic network |
| CN105516192B (en) * | 2016-01-14 | 2018-11-02 | 同盾控股有限公司 | A kind of mail address is safe to identify control method and device |
| CN107180030B (en) * | 2016-03-09 | 2020-11-17 | 创新先进技术有限公司 | Method and device for generating relational data on network |
| CN107203883B (en) * | 2016-03-17 | 2020-06-26 | 阿里巴巴集团控股有限公司 | Risk control method and equipment |
| CN107274042A (en) * | 2016-04-06 | 2017-10-20 | 阿里巴巴集团控股有限公司 | A kind of business participates in the Risk Identification Method and device of object |
| CN111784348A (en) * | 2016-04-26 | 2020-10-16 | 阿里巴巴集团控股有限公司 | Account risk identification method and device |
| CN106357657B (en) * | 2016-09-29 | 2020-02-21 | 恒大智慧科技有限公司 | Signing and issuing user management method and device |
| CN109600344B (en) * | 2017-09-30 | 2021-03-23 | 腾讯科技（深圳）有限公司 | Method and device for identifying risk group and electronic equipment |
| CN107871203A (en) * | 2017-09-30 | 2018-04-03 | 平安科技（深圳）有限公司 | Business personnel's behaviorist risk screens management method, application server and computer-readable recording medium |
| CN109697595B (en) * | 2017-10-20 | 2020-11-27 | 杭州海康威视系统技术有限公司 | Method and device for identifying attendance data, storage medium and computer equipment |
| CN107958341A (en) * | 2017-12-12 | 2018-04-24 | 阿里巴巴集团控股有限公司 | Risk Identification Method and device and electronic equipment |
| CN110009520B (en) * | 2018-01-02 | 2021-07-02 | 航天信息股份有限公司 | Abnormal grain transaction checking method and device |
| CN110278175B (en) * | 2018-03-14 | 2020-06-02 | 阿里巴巴集团控股有限公司 | Graph structure model training and garbage account identification method, device and equipment |
| CN108898418B (en) * | 2018-05-31 | 2023-06-23 | 康键信息技术（深圳）有限公司 | User account detection method, device, computer equipment and storage medium |
| CN109189805B (en) * | 2018-08-28 | 2021-01-01 | 拉扎斯网络科技（上海）有限公司 | Behavior processing method and apparatus, electronic device and computer-readable storage medium |
| CN109213857A (en) * | 2018-08-29 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of fraud recognition methods and device |
| CN109508879B (en) * | 2018-11-09 | 2023-07-18 | 创新先进技术有限公司 | Risk identification method, device and equipment |
| CN109829776B (en) * | 2018-12-14 | 2024-02-27 | 平安科技（深圳）有限公司 | Merchant risk assessment method, device, computer equipment and storage medium |
| CN110069923A (en) * | 2019-03-13 | 2019-07-30 | 咪咕文化科技有限公司 | A kind of method and relevant apparatus identifying risk subscribers |
| CN110084468B (en) * | 2019-03-14 | 2020-09-01 | 阿里巴巴集团控股有限公司 | Risk identification method and device |
| CN110415069A (en) * | 2019-06-20 | 2019-11-05 | 江苏常熟农村商业银行股份有限公司 | Order processing method, apparatus, system and storage medium based on financial business |
| CN112413832B (en) * | 2019-08-23 | 2021-11-30 | 珠海格力电器股份有限公司 | User identity recognition method based on user behavior and electric equipment thereof |
| CN110992135B (en) * | 2019-11-25 | 2024-01-23 | 拉扎斯网络科技（上海）有限公司 | Risk identification method and device, electronic equipment and storage medium |
| CN117611330B (en) * | 2024-01-23 | 2024-04-09 | 天津金城银行股份有限公司 | Credit data processing system, method, device, equipment and medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010150251A1 (en) * | 2009-06-22 | 2010-12-29 | United Parents Online Ltd. | Method and system of monitoring a network based communication among users |
| US20110029382A1 (en) * | 2009-07-30 | 2011-02-03 | Runu, Inc. | Automated Targeting of Information to a Website Visitor |
| US20110302097A1 (en) * | 2010-06-04 | 2011-12-08 | Microsoft Corporation | Behavior-based networks |
| CN102339445A (en) * | 2010-07-23 | 2012-02-01 | 阿里巴巴集团控股有限公司 | Method and system for evaluating credibility of network trade user |
| CN102413013B (en) * | 2011-11-21 | 2013-11-06 | 北京神州绿盟信息安全科技股份有限公司 | Method and device for detecting abnormal network behavior |
| CN102571484B (en) * | 2011-12-14 | 2014-08-27 | 上海交通大学 | Method for detecting and finding online water army |
-
2012
- 2012-08-03 CN CN201210275476.7A patent/CN103577991B/en active Active
- 2012-08-03 CN CN201910114022.3A patent/CN110009372B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| "C2C电子商务平台欺诈问题研究";陈鑫;《电子商务》;20110731;第22-23页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103577991B (en) | 2019-03-19 |
| CN110009372A (en) | 2019-07-12 |
| CN103577991A (en) | 2014-02-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110009372B (en) | User risk identification method and device | |
| CN108038696B (en) | Method and system for detecting bill swiping based on equipment identification code and social group information | |
| Wang et al. | Identify online store review spammers via social review graph | |
| CN103577988B (en) | A kind of method and apparatus for recognizing specific user | |
| CN108573432B (en) | Transaction supervision system and method for electronic commerce | |
| WO2017028735A1 (en) | Method and device for selecting and recommending display object | |
| US20090076868A1 (en) | Automated Electronic Commerce Data Analyzing and Sales System | |
| US8943060B2 (en) | Systems, methods and apparatus for identifying links among interactional digital data | |
| CN102903047A (en) | Method and device for releasing commodity information | |
| CN103853948A (en) | User identity recognizing and information filtering and searching method and server | |
| CN108648038B (en) | Credit frying and malicious evaluation identification method based on subgraph mining | |
| CN102004999A (en) | Behaviour revenue model based collusion group identification method in electronic commerce network | |
| CN107305665A (en) | It is a kind of to differentiate wash sale, prevent the single method and device of brush | |
| CN110414902B (en) | E-commerce platform system based on internet | |
| Carpineto et al. | Learning to detect and measure fake ecommerce websites in search-engine results | |
| Maranzato et al. | Fraud detection in reputation systems in e-markets using logistic regression and stepwise optimization | |
| Cherqi et al. | Analysis of hacking related trade in the darkweb | |
| Tu et al. | Online auction segmentation and effective selling strategy: Trust and information asymmetry perspectives | |
| Ghose et al. | Reputation premiums in electronic peer-to-peer markets: Analyzing textual feedback and network structure | |
| Fire et al. | Large-Scale Shill Bidder Detection in E-commerce | |
| CN104331823B (en) | It is determined that the method and device for the middle keyword reservation price that releases news | |
| Al-Fawaeer | Exploring the relationship between e-commerce adoption and business strategy: an applied study on the Jordanian telecommunication companies | |
| Chengzhang et al. | Detecting spamming stores by analyzing their suspicious behaviors | |
| JP2009163598A (en) | Stock price information collection and analysis system and server thereof | |
| CN108763233A (en) | The method and apparatus of the identification of doubtful fake products commodity and classification based on big data |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200924 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20200924 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |