Risk control methods, systems, equipment, media and program products for Internet marketing

Abstract

The present disclosure provides a risk control method, system, device, medium and program product for Internet marketing. The risk control method for Internet marketing is applied to a marketing platform. The risk control method for Internet marketing includes: extracting user characteristics and their business relationships based on a nebula graph database, determining the association between user characteristics of multiple users, and building a business graph model; through the business graph model, marking multiple users whose user characteristics meet similar conditions as a group of fan users; in response to the fan users meeting the risk control rules, setting a risk control identifier for each user in the fan users; in response to the presence of users with increased business relationships in the marketing platform, querying the risk control score corresponding to the user characteristics of the users with increased business relationships; the risk control score is positively correlated with the number of risk control identifiers. The present disclosure can identify the situation where the same user repeatedly obtains benefits through multiple accounts, avoiding the waste of operating and promotion funds.

Description

Technical Field

[0001] The present invention relates to the field of risk management and control, and in particular to a risk control method, system, equipment, medium and program product for Internet marketing. Background Art

[0002] In order to promote their products, companies have launched a large number of operational promotion activities. With the development of mobile applications, companies will conduct operational promotion activities based on application software. Some users will take the opportunity to gain benefits from it, causing irreparable financial losses to the company. The current risk control method can only intercept users with abnormal behavior based on equipment or orders, and the intercepted users can obtain benefits again through other accounts, resulting in a waste of funds for operational promotion activities and failing to achieve the expected promotion effect. Summary of the invention

[0003] The technical problem to be solved by the present invention is to overcome the defect in the prior art that it is impossible to identify that the same user repeatedly obtains benefits through multiple accounts, and to provide a risk control method, system, equipment, medium and program product for Internet marketing.

[0004] The present invention solves the above technical problems through the following technical solutions:

[0005] In the first aspect, a risk control method for Internet marketing is provided. The risk control method for Internet marketing is applied to the marketing platform. The risk control method for Internet marketing includes:

[0006] Based on the nebula graph database, we extract user features and their business relationships, determine the associations between user features of multiple users, and build a business graph model. The business graph model uses businesses and users participating in the business as nodes, and the business relationships between users and businesses as edges.

[0007] Through the business graph model, multiple users whose user characteristics meet similar conditions are marked as a group of similar users;

[0008] In response to the fan user meeting the risk control rule, setting a risk control flag for each of the fan users;

[0009] In response to users with business relationship changes in the marketing platform, the risk control score corresponding to the user characteristics of the users with the business relationship changes is queried; the risk control score is positively correlated with the number of risk control identifiers.

[0010] Optionally, the risk control rule includes: the number of at least one type of business relationship held by at least one user among the same users is greater than the risk control threshold corresponding to the at least one type of business relationship;

[0011] In response to the fan user meeting the risk control rule, the step of setting a risk control flag for each of the fan users includes:

[0012] Risk control identification is set for users who hold at least one type of business relationship with a number greater than the corresponding risk control threshold and for the remaining users among the same users.

[0013] Optionally, the risk control rule includes: the sum of the number of at least one type of business relationships held by users among the same users is greater than the risk control threshold corresponding to the at least one type of business relationship.

[0014] Optionally, the user features include: primary features and secondary features, the primary features including a unique identifier of the user identity;

[0015] The similarity condition includes: the number of identical primary features in the user features of multiple different users is greater than a first quantity threshold, and / or the number of identical secondary features in the user features of multiple different users is greater than a second quantity threshold.

[0016] Optionally, after the step of querying and returning the risk control score corresponding to the user feature of the user, the risk control method for Internet marketing further includes:

[0017] Divide the user's target risk level according to the risk control score;

[0018] Based on the correspondence between the preset risk level and the risk control strategy, the risk control strategy corresponding to the preset risk level that is the same as the target risk level is executed for the user.

[0019] Optionally, the step of executing a risk control strategy corresponding to a preset risk level that is the same as the target risk level on the user includes:

[0020] In response to the user characteristics of the user not being included in the whitelist, a risk control strategy corresponding to a preset risk level that is the same as the target risk level is executed on the user.

[0021] In the second aspect, a risk control system for Internet marketing is provided. The risk control system for Internet marketing is applied to a marketing platform. The risk control system for Internet marketing includes:

[0022] The construction module is used to extract user features and their business relationships based on the nebula graph database, determine the association between user features of multiple users, and build a business graph model; the business graph model uses businesses and users participating in the business as nodes, and the business relationships between users and businesses as edges;

[0023] A marking module is used to mark multiple users whose user characteristics meet similar conditions as a group of similar users through a business graph model;

[0024] A risk control identification setting module, configured to set a risk control identification for each fan user in response to the fan user meeting the risk control rules;

[0025] The query module is used to query the risk control score corresponding to the user characteristics of the user with the changed business relationship in response to the user with the changed business relationship in the marketing platform; the risk control score is positively correlated with the number of risk control identifiers.

[0026] According to a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, wherein when the processor executes the computer program, any of the above-mentioned Internet marketing risk control methods is implemented.

[0027] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, any of the above-mentioned Internet marketing risk control methods is implemented.

[0028] In a fifth aspect, a computer program product is provided, comprising a computer program, characterized in that when the computer program is executed by a processor, it implements any of the above-mentioned Internet marketing risk control methods.

[0029] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present disclosure.

[0030] The positive and progressive effects of the present disclosure are as follows: the present disclosure can intuitively reflect the potential relationship between users through the business graph model, mark multiple users whose user characteristics meet similar conditions as a group of fan users, thereby identifying different accounts of the same user, and when a group of fan users meets the risk control rules, it is possible to set a risk control identification for each user in the group of fan users to obtain a risk control score; when the business relationship held by the user changes, for example, when the user participates in a marketing activity, the risk control score corresponding to the user characteristic is returned, and users with higher risk control scores can be identified from the users participating in the marketing activity, thereby effectively identifying the situation where the same user repeatedly obtains benefits through multiple accounts, thereby avoiding the waste of funds for operating and promotion activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A flow chart of a risk control method for Internet marketing provided by an exemplary embodiment of the present disclosure;

[0032] Figure 2 Another flow chart of a risk control method for Internet marketing provided by an exemplary embodiment of the present disclosure;

[0033] Figure 3 A module schematic diagram of a risk control system for Internet marketing provided by an exemplary embodiment of the present disclosure;

[0034] Figure 4 The present invention provides a schematic structural diagram of an electronic device according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0035] The present disclosure is further described below by way of examples, but the present disclosure is not limited to the scope of the examples.

[0036] Prefixes such as "first" and "second" are used in the embodiments of the present disclosure only to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. The use of prefixes such as ordinal numbers to distinguish description objects in the embodiments of the present disclosure does not constitute a limitation on the described objects. For the statement of the described objects, please refer to the description in the context of the claims or embodiments, and no unnecessary limitation should be constituted due to the use of such prefixes. In addition, in the description of the present embodiment, unless otherwise specified, the meaning of "plurality" is two or more.

[0037] In the embodiments of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0038] Example 1

[0039] Figure 1 A flow chart of a risk control method for Internet marketing provided by an exemplary embodiment of the present disclosure, the risk control method for Internet marketing is applied to a marketing platform, and the risk control method for Internet marketing includes:

[0040] Step S101: extract user features and their business relationships based on the nebula graph database, determine the association between user features of multiple users, and build a business graph model.

[0041] The business graph model uses businesses and users involved in the business as nodes, and the business relationships between users and businesses as edges.

[0042] It should be noted that user characteristics and business relationships are obtained from the marketing platform. When the user completes the login registration, the marketing platform can collect the user characteristics and business relationships of the user. User characteristics include but are not limited to ID card, device ID, IP address, user ID, mobile phone number, email address, etc., which can be set according to the actual situation and are not specifically limited here.

[0043] Business includes but is not limited to basic behaviors, order behaviors, member benefits and marketing activities. Basic behaviors include but are not limited to login and registration, order behaviors include but are not limited to creating orders, paying orders and refunding orders, member benefits include but are not limited to opening a membership and receiving benefits, and marketing activities include but are not limited to cash back and issuing coupons.

[0044] Each business contains business information corresponding to the business, including but not limited to order number, activity number, and equity number. The business can be set according to actual conditions and is not specifically limited here.

[0045] Business relationships include, but are not limited to, basic behavioral relationships, order relationships, membership relationships, and marketing activity relationships, which can be set up based on actual conditions and are not specifically limited here.

[0046] The following are examples of basic behavioral relationships, order relationships, membership relationships, and marketing activity relationships.

[0047] When a user registers on a marketing platform, the user's business relationship includes a basic behavioral relationship. When a user logs in and registers on a marketing platform, the basic behavioral relationship is bound based on the user ID, device ID, and mobile phone number in the user characteristics. The basic behavioral relationship includes login and registration, and may also include placing an order, which can be adjusted according to actual conditions and is not specifically limited here. Specifically, by binding the user ID to the basic behavioral relationship with other user characteristics, the user ID can be used as a criterion for subsequent judgment of fan users. In response to the fan user meeting the risk control rules, a risk control identifier is set at the user ID of each user among the fan users.

[0048] When users create orders, pay orders, and refund orders on the marketing platform, an order relationship is established between user characteristics (including but not limited to user ID, device ID, mobile phone number, contact person, passenger, ID card) and the order number in the business information.

[0049] When a user opens a membership on the marketing platform, a membership relationship is established between the user's characteristics (including but not limited to user ID, device ID, mobile phone number, ID card) and the rights and interests number in the business information.

[0050] When a user participates in a marketing activity, a marketing activity relationship is established between the user characteristics (including but not limited to user ID, device ID, mobile phone number, ID card) and the activity number in the business information.

[0051] By analyzing user data to build a business graph model, the business graph model can be used to reflect the user's operational behavior on the marketing platform and the relationship between multiple users. User data includes but is not limited to user transaction data, user login and registration data, user order information, etc. When a user repeatedly changes his or her identity information, such as mobile phone number, ID card, and user ID, to bypass the number of redemption restrictions to obtain activity rewards, the relationship between the user characteristics of multiple users can be determined, allowing for timely interception to avoid economic losses.

[0052] When a user registers a virtual account and repeatedly receives rewards, for example, if there is a new customer rebate activity on a marketing platform, newly registered users can get a 5 yuan rebate. Users can repeatedly register new accounts and repeatedly receive the 5 yuan rebate through virtual accounts or virtual mobile phone numbers. Since users have the same device ID, IP information, and ID card, the business graph model can be used to mark multiple users whose user characteristics meet similar conditions as a group of similar users.

[0053] When users obtain high subsidies at low prices, such as opening a membership for 15 yuan, members are entitled to receive VIP coupons when placing orders. Some users repeatedly receive VIP coupons by placing orders far below the value of the VIP coupons and resell them in the market. Through the business graph model, we can analyze the business relationship between users and businesses to determine whether it is necessary to set a risk control mark for the user.

[0054] Step S102: Using the business graph model, mark multiple users whose user characteristics meet similar conditions as a group of like-minded users.

[0055] Step S103: In response to the fan users meeting the risk control rules, a risk control flag is set for each fan user.

[0056] The number of risk control rules can be set according to actual conditions and is not specifically limited here.

[0057] In one embodiment, the graying threshold, graying depth and expiration time can be configured. Graying means setting a risk control flag for each fan user who meets similar conditions. Graying a user indicates that the user has certain risks. If the number of at least one type of business relationship held by at least one fan user is greater than the graying threshold corresponding to at least one type of business relationship, each fan user will be grayed.

[0058] The gray threshold can be the same as the risk control threshold, or it can be different from the risk control threshold. The gray threshold, gray depth, and expiration time can be set according to actual conditions and are not specifically limited here.

[0059] The expiration time is used to ensure the timeliness of the data. After the expiration time, the data may no longer reflect the user's latest risk status. The graying depth refers to the degree of association between the user characteristics that need to be marked with risk control marks and the user ID. Specifically, when a user ID is subject to risk control, whether other user characteristics of the user of the user ID need to be marked with risk control marks, and determine the risk level of the risk control marks that need to be marked.

[0060] Step S104: In response to users with changed business relationships in the marketing platform, query the risk control score corresponding to the user characteristics of the users with changed business relationships; the risk control score is positively correlated with the number of risk control identifiers.

[0061] Based on R=LCE (risk control score=cost*consequence*control), the benchmark score of the corresponding abnormal feature is calculated, and the risk control score is calculated by accumulation. Among them, R represents the risk score of a single risk control identifier, L represents the possibility of risk control, E represents the consequences and losses caused, and C represents the effectiveness of the control measures. The R corresponding to multiple risk control identifiers are added up to obtain the benchmark score of the abnormal feature. It can be understood that the more risk control identifiers a user has, the higher the risk control score, which means that the user is at greater risk and the user may have abnormal operations such as repeatedly obtaining benefits.

[0062] The coefficient of possibility L is determined according to the user cost. When the user cost required for the user to complete the operation that triggers risk control is at the first level, the coefficient of possibility L is 1. The operations with user cost at the first level include but are not limited to operations that can be completed by simple repetition such as brushing orders. When the user cost required for the user to complete the operation that triggers risk control is at the second level, the coefficient of possibility L is 0.75. The operations with user cost at the second level include but are not limited to operations or behaviors that bypass marketing activity rules. When the user cost required for the user to complete the operation that triggers risk control is at the third level, the coefficient of possibility L is 0.25. The operations with user cost at the third level include but are not limited to script writing based on certain development literacy. When the user cost required for the user to complete the operation that triggers risk control is at the fourth level, the coefficient of possibility L is 0.1. The operations with user cost at the fourth level include but are not limited to professional teams moving orders. Moving orders specifically refers to the behavior of some agents or individuals who, without the consent of the hotel, arbitrarily put hotel rooms on the platform for sale. They earn the difference by moving room information between different platforms and adding prices. The user costs from the first level to the fourth level gradually increase. The above values ​​are only for example, and can be selected according to actual conditions, and are not specifically limited here.

[0063] When the user behavior has no impact on funds and resources, the coefficient of consequence and loss E is 0; when the user behavior does not cause financial loss but occupies resources (such as crawlers, order brushing and continued occupation, etc.), the coefficient of consequence and loss E is 1; when the user behavior causes indirect financial loss (such as cash back, coupon collection and order transfer, etc.), the coefficient of consequence and loss E is 2; when the user behavior is illegal or causes direct financial loss (such as withdrawal, coupon use and rights and interests, etc.), the coefficient of consequence and loss E is 3. The above values ​​are for example only, and can be selected according to actual conditions, and are not particularly limited here.

[0064] When user behavior cannot be intercepted due to objective reasons (for example, users sell membership rights in the secondary market), the coefficient of the effectiveness of the control measures C is 5. For user behavior that triggers risk control, if a solution can be launched in a short period of time (more than one day), the coefficient of the effectiveness of the control measures C is 4; if it can be intercepted within one day, the coefficient of the effectiveness of the control measures C is 2-3; if it can be intercepted immediately, the coefficient of the effectiveness of the control measures C is 1.

[0065] Based on the value of risk score R, users can be divided into four levels: no risk, low risk, medium risk and high risk. When the value of R is 0, the corresponding risk level is no risk, when the value of R is in [1,3), the corresponding risk level is low risk, when the value of R is in [3,9), the corresponding risk level is medium risk, and when the value of R is in [9,15), the corresponding risk level is high risk.

[0066] The benchmark score can be used to determine the risk level of the user. The risk level of a user with a benchmark score of 0 and no risk control identifier is no risk; the risk level of a user with a benchmark score of 1 to 10 and only low-risk risk control identifiers is low risk; the risk level of a user with a benchmark score of 10 to 20 and only low-risk and medium-risk risk control identifiers is medium risk; the risk level of a user with a benchmark score of 30 or more and high-risk risk control identifiers is high risk. The above values ​​are for example only and can be selected according to actual conditions and are not specifically limited here.

[0067] Specifically, the marketing platform sets up a risk control identification query interface. When a user's business relationship increases, that is, the user participates in order behavior, membership rights and marketing activities, the risk control identification query interface is used to query the risk control score corresponding to the user characteristics of the user whose business relationship has increased, and use the risk control score as the basis for whether to implement risk control strategies for the user in the future.

[0068] The increase of a user's business relationship includes but is not limited to the user's participation in order behavior, membership benefits and marketing activities. For example, the number of a user's business relationships is 3. When the user participates in a marketing activity, a business relationship between the user's characteristics and the activity number is established. At this time, the number of the user's business relationships changes from 3 to 4.

[0069] In this embodiment, the potential relationship between users can be intuitively reflected through the business graph model, and multiple users whose user characteristics meet similar conditions can be marked as a group of fan users, so as to identify different accounts of the same user. When a group of fan users meets the risk control rules, a risk control identification can be set for each user in the group of fan users to obtain a risk control score. When the business relationships held by the user increase, for example, when the user participates in a marketing activity, the risk control score corresponding to the user characteristic is returned, and users with higher risk control scores can be identified from the users participating in the marketing activity, thereby effectively identifying the situation where the same user repeatedly obtains benefits through multiple accounts, thereby avoiding the waste of funds for operating promotion activities.

[0070] In one embodiment, in response to the presence of a user with reduced business relationship in the marketing platform, the risk control score corresponding to the user characteristics of the user with reduced business relationship is queried. The reduction of business relationship includes but is not limited to canceling an order, which can be set according to actual conditions and is not particularly limited here.

[0071] In this embodiment, when the business relationship decreases, the risk control score of the user can be queried to determine the risk level of the user and avoid wasting funds for operating promotion activities.

[0072] In one embodiment, different risk control identifiers correspond to different identifier scores, and the risk control score is obtained by accumulating the identifier scores.

[0073] The identification score can be set according to actual conditions and is not particularly limited here.

[0074] In this embodiment, the risk control score is obtained by cumulative calculation, which has a small amount of calculation, is simple and easy to implement, and can quickly and timely obtain the risk control score to determine whether the user repeatedly obtains benefits through multiple accounts.

[0075] In one embodiment, the risk control rule includes: among the same users, there is at least one user whose number of at least one type of business relationship is greater than the risk control threshold corresponding to each type of business relationship.

[0076] Step S103 includes: setting risk control flags for users who hold at least one type of business relationship with a number greater than a corresponding risk control threshold and for the remaining users among the same users.

[0077] Specifically, a risk control threshold is set for each business relationship, for example, cashback per person is limited to n times per day, and the duration of the risk control strategy is configured to reduce the situation where the operation promotion benefits are repeatedly obtained by the same users. The risk control threshold can be set according to the actual situation and is not specifically limited here.

[0078] Taking the risk control threshold corresponding to the marketing activity as 3 as an example, user account A, user account B and user account C are marked as fan users. If user account B participates in more than three marketing activities, a risk control flag is set for each fan user, that is, a risk control flag is set for user account A, user account B and user account C.

[0079] In this embodiment, when there is at least one user among the fan users whose number of at least one type of business relationships is greater than the corresponding risk control threshold, not only is a risk control flag set for the user whose number of at least one type of business relationships is greater than the corresponding risk control threshold, but risk control flags are also set for the remaining users among the fan users to avoid wasting funds for operating and promotion activities.

[0080] In one embodiment, fan users complying with risk control rules includes: the sum of the number of at least one type of business relationships held by users among the fan users is greater than the risk control threshold corresponding to the at least one type of business relationship.

[0081] Taking the risk control threshold corresponding to marketing activities as 3, and the fan users including five users as an example, assuming that there is a user among the fan users who participated in 2 marketing activities, and the rest of the fan users participated in 1 marketing activity, then the sum of the number of at least one type of business relationship held by the users among the fan users is 2+1+1+1+1=6>3, which has exceeded the risk control threshold, and it is necessary to set a risk control flag for each user among the fan users.

[0082] In this embodiment, when the sum of the number of at least one type of business relationships held by users among the fan users is greater than the corresponding risk control threshold, it can effectively identify the situation where the fan users repeatedly obtain benefits through multiple accounts, thereby avoiding the waste of funds for operating and promotion activities.

[0083] In one embodiment, the user features include: primary features, the primary features include a unique identifier of the user identity. The similarity condition includes: the number of the same primary features in the user features of multiple different users is greater than a first number threshold.

[0084] The first quantity threshold can be selected according to actual conditions. The primary features include but are not limited to identity cards and passports, etc., which can be selected according to actual conditions and are not particularly limited here.

[0085] Specifically, assuming that the primary feature only includes the ID card, the first quantity threshold is 1, and the similarity condition is met, that is, the ID cards are the same. When the ID cards of multiple users are the same, it means that the multiple users are used by the same person. Multiple users whose user features meet the similarity condition are marked as a group of same-person users.

[0086] In this embodiment, multiple users are marked as a group of fan users based on the number of their same primary features, thereby identifying different accounts used by the same person. If a fan user repeatedly obtains benefits, a risk control flag can be set for each fan user.

[0087] In one embodiment, the user features further include: secondary features, and the similarity condition includes: the number of identical secondary features in the user features of multiple different users is greater than a second number threshold.

[0088] The second quantity threshold can be selected according to actual conditions, and the secondary features include but are not limited to mobile phone number, user ID, and email address. The specific features can be selected according to actual conditions and are not particularly limited here.

[0089] Specifically, assuming that the first quantity threshold is 2, the similarity condition is met, that is, there are more than two identical secondary features in the user characteristics of multiple users, for example, the mobile phone number, user ID and email address are all the same. When there are more than two identical secondary features in the user characteristics of multiple users, it means that the multiple users are used by the same person, and the multiple users whose user characteristics meet the similarity condition are marked as a group of same users.

[0090] In this embodiment, multiple users are marked as a group of fan users based on the number of their same secondary features, thereby identifying different accounts used by the same person. If a fan user repeatedly obtains benefits, a risk control flag can be set for each fan user.

[0091] In one embodiment, the first quantity threshold, the second quantity threshold and the third quantity threshold are obtained by analyzing historical abnormal data. Specifically, the appropriate first quantity threshold, the second quantity threshold and the third quantity threshold are determined by analyzing the value distribution of the historical abnormal data.

[0092] In one embodiment, the user features further include: tertiary features, and the similarity condition includes: the number of identical tertiary features in the user features of multiple different users is greater than a third number threshold.

[0093] The third quantity threshold can be selected according to actual conditions. The third-level features include but are not limited to device ID and IP. The specific features can be selected according to actual conditions and are not particularly limited here.

[0094] Specifically, assuming that the first quantity threshold is 3, the similarity condition is met, that is, there are more than 3 identical tertiary features in the user features of multiple users. When there are more than 3 identical secondary features in the user features of multiple users, it means that the multiple users are used by the same person, and the multiple users whose user features meet the similarity condition are marked as a group of same users.

[0095] In this embodiment, multiple users are marked as a group of fan users based on the number of their same tertiary features, thereby identifying different accounts used by the same person. If a fan user repeatedly obtains benefits, a risk control identification can be set for each fan user.

[0096] In one embodiment, after step S104, the risk control method for Internet marketing further includes: dividing the target risk level of the user according to the risk control score. Based on the correspondence between the preset risk level and the risk control strategy, the risk control strategy corresponding to the preset risk level that is the same as the target risk level is executed on the user.

[0097] Assuming that the risk control score ranges from 0 to 10, 0 to 10 can be divided into multiple intervals, for example, 0 to 2, 3 to 4, 5 to 6, 7 to 8, and 9 to 10. The risk level of users with a risk control score in the range of 0 to 2 is E, the risk level of users with a risk control score in the range of 3 to 4 is D, and so on. The risk level of users with a risk control score in the range of 9 to 10 is A.

[0098] The above-mentioned risk control score value range and division interval are only examples and can be set according to actual conditions and are not particularly limited here.

[0099] Risk control strategies can be set in advance. Specifically, risk control thresholds can be set for each business relationship. For example, cash back per person should not exceed n times per day. The duration of the risk control strategy can be configured to reduce the situation where operational promotion benefits are repeatedly obtained by same users.

[0100] Risk control strategies include but are not limited to interception and warning, which can be selected according to actual conditions and are not specifically limited here. The duration of the risk control strategy and the correspondence between the preset risk level and the risk control strategy can be set according to actual conditions and are not specifically limited here.

[0101] In this embodiment, the target risk level is used to guide the selection of risk control strategies and quantify the risks faced by each user, thereby more effectively controlling risks and avoiding wasting funds for operating promotion activities.

[0102] In one embodiment, user characteristics include but are not limited to IP addresses. Traffic is monitored on the IP address. When the traffic increase of the IP address is greater than the traffic threshold, it indicates that a traffic surge has occurred at the IP address, and an alarm is issued to the user whose user characteristics include the IP address.

[0103] The flow threshold can be set according to actual conditions and is not specifically limited here.

[0104] In this embodiment, the traffic of the IP address is monitored, and when a traffic surge occurs, an alarm is issued to the user, thereby serving as a warning to the user.

[0105] In one embodiment, the step of executing a risk control strategy corresponding to a preset risk level that is the same as the target risk level on the user includes:

[0106] In response to the user characteristics of the user not being included in the whitelist, a risk control strategy corresponding to a preset risk level that is the same as the target risk level is executed on the user.

[0107] The whitelist includes user characteristics of users who do not need to execute risk control strategies. The user characteristics in the whitelist can be displayed, added, edited, and deleted as needed. The user characteristics of users whose risk control scores meet the safety range can also be added to the whitelist. The specific settings can be made according to actual conditions and are not particularly limited here.

[0108] In this embodiment, if the user characteristics are not in the whitelist, the risk control strategy is executed on the user. If the user characteristics are in the whitelist, there is no need to execute the risk control strategy, thereby reducing unnecessary review of users in the whitelist and improving risk control efficiency and accuracy.

[0109] In one embodiment, the risk control list is displayed, added, edited, and deleted through the risk control list management page, where the risk control list includes a blacklist and a whitelist.

[0110] The user characteristics in the blacklist will be restricted in some behaviors, and will enter the algorithm model for abstract modeling, and abnormal behaviors will be analyzed based on machine learning, while the user characteristics in the whitelist will not be restricted by risk control rules.

[0111] In this embodiment, the risk control list is managed through the risk control list management page. The risk control list management page can be flexibly set according to business needs, thereby improving the flexibility of risk control and facilitating overall regulation.

[0112] See also Figure 2 , further explain the process of risk control methods for Internet marketing. Figure 2 Qmq transmission refers to the data transmission process of a self-developed message middleware Qunar Messaging Queue (abbreviated as qmq). Other transmission methods can also be used in actual use, which is not particularly limited here.

[0113] Based on the nebula graph database, we extract user features and their business relationships, determine the associations between user features of multiple users, and build a business graph model. The business graph model uses businesses and users participating in the business as nodes, and the business relationships between users and businesses as edges. After building the business graph model, we set risk control rules and graying rules.

[0114] When a user triggers an action such as placing an order, receiving cashback, or participating in an event, the business relationship held by the user increases. The risk control score corresponding to the user characteristics of the user with the increased business relationship is queried to complete the risk verification. If the risk control score reaches the preset interception standard, the user is intercepted. If the risk control verification passes, it means that the user has not yet reached the preset interception standard.

[0115] The business graph model is used to determine whether there are any fan users and to verify the risk control rules. If the user hits the risk control rules, that is, meets the risk control rules, the user who hits the risk control rules is a potential user and has a certain degree of risk. The potential user will be grayed out to avoid wasting funds for operation and promotion activities. If the user does not hit the risk control rules, the risk control data is collected, the user's high-risk behavior is analyzed, and the risk control rules and graying rules are adjusted.

[0116] Adjustments to risk control rules include but are not limited to adjusting risk control thresholds and the duration of risk control strategies. Adjustments to graying rules include but are not limited to adjusting graying depth, graying thresholds and expiration time. The specific adjustments can be made based on actual conditions and are not specifically limited here.

[0117] Example 2

[0118] Corresponding to the aforementioned embodiments of the risk control method for Internet marketing, the present disclosure also provides embodiments of the risk control system for Internet marketing.

[0119] Figure 3 A schematic diagram of a module of a risk control system for Internet marketing provided by an exemplary embodiment of the present disclosure is provided. The risk control system for Internet marketing is applied to a marketing platform. The risk control system for Internet marketing includes:

[0120] The construction module is used to extract user features and their business relationships based on the nebula graph database, determine the association between user features of multiple users, and build a business graph model; the business graph model uses businesses and users participating in the business as nodes, and the business relationships between users and businesses as edges;

[0121] A marking module is used to mark multiple users whose user characteristics meet similar conditions as a group of similar users through a business graph model;

[0122] A risk control identification setting module, configured to set a risk control identification for each fan user in response to the fan user meeting the risk control rules;

[0123] The query module is used to query the risk control score corresponding to the user characteristics of the user with added business relationship in response to the user with added business relationship in the marketing platform; the risk control score is positively correlated with the number of risk control identifiers.

[0124] Optionally, the risk control rule includes: the number of at least one type of business relationship held by at least one user among the same users is greater than the risk control threshold corresponding to the at least one type of business relationship;

[0125] The risk control identification setting module includes:

[0126] The identification setting unit is used to set risk control identification for users who hold at least one type of business relationship with a number greater than the corresponding risk control threshold and the remaining users among the same users.

[0127] Optionally, the risk control rule includes: the sum of the number of at least one type of business relationships held by users among the same users is greater than the risk control threshold corresponding to the at least one type of business relationship.

[0128] Optionally, the user features include: primary features and secondary features, the primary features including a unique identifier of the user identity;

[0129] The similarity condition includes: the number of identical primary features in the user features of multiple different users is greater than a first quantity threshold, and / or the number of identical secondary features in the user features of multiple different users is greater than a second quantity threshold.

[0130] Optionally, the risk control system for Internet marketing also includes:

[0131] The risk level classification module is used to classify the user's target risk level according to the risk control score;

[0132] The strategy execution module is used to execute the risk control strategy corresponding to the preset risk level that is the same as the target risk level for the user based on the correspondence between the preset risk level and the risk control strategy.

[0133] Optionally, the policy execution module includes:

[0134] The policy execution unit is used to execute a risk control policy corresponding to a preset risk level that is the same as a target risk level on the user in response to the user characteristics not being included in the whitelist.

[0135] The business graph model can intuitively reflect the potential relationship between users, and multiple users whose user characteristics meet similar conditions can be marked as a group of fan users, so as to identify different accounts of the same user. When a group of fan users meets the risk control rules, a risk control mark can be set for each user in the group to obtain a risk control score. When the number of business relationships held by users increases, for example, when users participate in a marketing activity, the risk control score corresponding to the user characteristics is returned, and users with higher risk control scores can be identified from among the users participating in the marketing activity, thereby effectively identifying the situation where the same user repeatedly obtains benefits through multiple accounts, avoiding the waste of funds for operating promotion activities.

[0136] As for the system embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The system embodiment described above is only illustrative, in which the units described as separate components may or may not be physically separated, and the components as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution.

[0137] Example 3

[0138] Figure 4 This is a schematic diagram of the structure of an electronic device shown in an example embodiment of the present disclosure, wherein the electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor, and when the processor executes the computer program, the risk control method for Internet marketing of any of the above embodiments is implemented. Figure 4 The electronic device 90 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0139] like Figure 4 As shown, the electronic device 90 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 90 may include, but are not limited to: at least one processor 91, at least one memory 92, and a bus 93 connecting different system components (including the memory 92 and the processor 91).

[0140] The bus 93 includes a data bus, an address bus, and a control bus.

[0141] The memory 92 may include a volatile memory, such as a random access memory (RAM) 921 and / or a cache memory 922 , and may further include a read-only memory (ROM) 923 .

[0142] The memory 92 may also include a program tool 925 (or utility) having a set (at least one) of program modules 924, such program modules 924 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0143] The processor 91 executes various functional applications and data processing by running the computer program stored in the memory 92, such as the risk control method for Internet marketing provided in any of the above embodiments.

[0144] The electronic device 90 may also communicate with one or more external devices 94 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 95. Furthermore, the electronic device 90 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 96. As shown, the network adapter 96 communicates with other modules of the electronic device 90 via a bus 93. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 90, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0145] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0146] Example 4

[0147] The embodiments of the present disclosure also provide a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the risk control method for Internet marketing provided by any of the above embodiments is implemented.

[0148] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0149] Example 5

[0150] The embodiments of the present disclosure also provide a computer program product, including a computer program, which implements any of the above-mentioned Internet marketing risk control methods when executed by a processor.

[0151] Among them, the program code for executing the computer program product of the present disclosure can be written in any combination of one or more programming languages, and the program code can be executed completely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or completely on the remote device.

[0152] Although the specific embodiments of the present disclosure are described above, those skilled in the art should understand that this is only an example, and the protection scope of the present disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, but these changes and modifications all fall within the protection scope of the present disclosure.

Claims

1. A risk control method for Internet marketing, characterized in that: The risk control method for Internet marketing is applied to the marketing platform, and the risk control method for Internet marketing includes: Based on the nebula graph database, feature extraction is performed on user features and their business relationships, the association between the user features of multiple users is determined, and a business graph model is constructed; the business graph model uses businesses and users participating in the business as nodes, and the business relationships between the users and the businesses as edges; By using the business graph model, marking the multiple users whose user characteristics meet similar conditions as a group of like users; In response to the fan users meeting the risk control rules, setting a risk control identifier for each of the fan users; In response to the presence of the user with the changed business relationship in the marketing platform, the risk control score corresponding to the user feature of the user with the changed business relationship is queried; the risk control score is positively correlated with the number of the risk control identifiers.

2. The risk control method for Internet marketing according to claim 1, characterized in that: The risk control rule includes: the number of at least one type of business relationships held by at least one user among the same users is greater than the risk control threshold corresponding to the at least one type of business relationship; In response to the fan user meeting the risk control rule, the step of setting a risk control identifier for each of the fan users includes: Risk control identification is set for users who hold at least one type of business relationship with a number greater than the corresponding risk control threshold and for the remaining users among the same users.

3. The risk control method for Internet marketing according to claim 1, characterized in that: The risk control rule includes: the sum of the number of at least one type of business relationships held by the users among the fellow users is greater than the risk control threshold corresponding to the at least one type of business relationships.

4. The risk control method for Internet marketing according to claim 1, characterized in that: The user characteristics include: primary characteristics and secondary characteristics, the primary characteristics include a unique identifier of the user identity; The similarity condition includes: the number of identical primary features in the user features of multiple different users is greater than a first quantity threshold, and / or the number of identical secondary features in the user features of multiple different users is greater than a second quantity threshold.

5. The risk control method for Internet marketing according to claim 1, characterized in that: After the step of querying and returning the risk control score corresponding to the user's user characteristics, the Internet marketing risk control method further includes: Dividing the target risk level of the user according to the risk control score; Based on the correspondence between the preset risk level and the risk control strategy, the risk control strategy corresponding to the preset risk level that is the same as the target risk level is executed for the user.

6. The risk control method for Internet marketing according to claim 5, characterized in that: The step of executing a risk control strategy corresponding to a preset risk level that is the same as the target risk level on the user includes: In response to the whitelist not including the user characteristics of the user, a risk control strategy corresponding to a preset risk level that is the same as the target risk level is executed on the user.

7. A risk control system for Internet marketing, characterized in that: The risk control system of Internet marketing is applied to the marketing platform. The risk control system of Internet marketing includes: A construction module is used to extract user features and their business relationships based on the nebula graph database, determine the association between user features of multiple users, and build a business graph model; the business graph model uses businesses and users participating in the business as nodes, and the business relationships between the users and the businesses as edges; A marking module, used to mark multiple users whose user characteristics meet similar conditions as a group of similar users through the business graph model; A risk control identification setting module, configured to set a risk control identification for each of the fan users in response to the fan users meeting the risk control rules; The query module is used to query the risk control score corresponding to the user characteristics of the user with changed business relationship in response to the user with changed business relationship in the marketing platform; the risk control score is positively correlated with the number of risk control identifiers.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, characterized in that: When the processor executes the computer program, the risk control method for Internet marketing described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the risk control method for Internet marketing described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the risk control method for Internet marketing as described in any one of claims 1 to 6 is implemented.

Images