Requity blind box generation and sending method and device and electronic equipment

Abstract

The invention provides a right blind box generation and sending method and device and electronic equipment, and relates to the technical field of computers. The method applied to the right and interest platform server comprises the following steps: based on attribute vectors of rights and interests in a right and interest pool, carrying out dynamic clustering division on the rights and interests to form a plurality of right and interest sub-pools; in response to the blind box acquisition instruction, a right blind box is generated in a layered progressive mode, right information in the right blind box is encrypted, and a zero-knowledge proof used for proving that the right information meets a predetermined condition is attached to the right blind box; and in response to a blind box unlocking instruction, verifying the validity of the zero-knowledge proof, and after the verification is passed, sending the right blind box to the user terminal equipment. The layered progressive technology and the zero-knowledge proof technology are adopted, the right center server is assisted to generate the right blind box, the right blind box generation mode can improve the surprise and experience of the user, and the fairness and transparency of right distribution are improved.

Description

Technical Field

[0001] The application generally relates to the field of computer technology, and in particular to a method, apparatus and electronic device for generating and sending rights blind boxes. Background Technology

[0002] With fierce competition among internet benefit platforms and the growing demand for personalized services, VIP user benefits have become a crucial means for platforms to attract and retain high-value users. Currently, the industry primarily utilizes expert-defined rules to provide benefit distribution capabilities, such as the following VIP benefit distribution methods: 1) Regularly distribute fixed benefit packages: such as giving VIP users a fixed amount of vouchers or points every month; 2) Set up a VIP exclusive redemption mall: Provide a selection of special offers and services prepared exclusively for VIP users for them to choose from; 3) Offer VIP-exclusive activities: such as exclusive discounts and lucky draws for VIP users; 4) Provide VIP exclusive customer service channel: Provide VIP users with priority response, dedicated service and other thoughtful services.

[0003] However, the traditional method of distributing rights through the rights center has the following problems: 1) The benefits are highly homogenized, lacking surprise and engagement, making it difficult to provide an attractive and exclusive experience; 2) The redemption process is open and transparent, but lacks privacy protection; 3) The value of rights and interests is difficult to be effectively conveyed, and users do not perceive it strongly. Summary of the Invention

[0004] This application provides a method for generating and sending benefit blind boxes to address some of the deficiencies mentioned in the background technology.

[0005] In a first aspect, embodiments of this application provide a method for generating and sending rights blind boxes, applied to a rights platform server, including: Based on the attribute vectors of the rights in the rights pool, the rights are dynamically clustered and divided into multiple rights sub-pools. The attribute vectors are ordered sequences formed by quantified values ​​corresponding to one or more predefined rights attributes arranged in a fixed order. In response to the blind box acquisition command, rights blind boxes are generated in a hierarchical and progressive manner. The rights information in the rights blind boxes is encrypted and accompanied by zero-knowledge proofs to prove that the rights information meets predetermined conditions. The hierarchical and progressive manner is to select rights from each rights sub-pool in descending order of weight until a rights blind box is generated. In response to the blind box unlocking command, the validity of the zero-knowledge proof is verified, and after successful verification, the rights blind box is sent to the user terminal device.

[0006] In a second aspect, embodiments of this application provide a method for generating and sending benefit blind boxes, applied to a user terminal device, comprising: sending a benefit blind box acquisition request to a benefit platform server; receiving a benefit blind box from the benefit platform server, wherein the benefit blind box is obtained based on the benefit blind box generation and sending method described in the first aspect; and after sending a benefit blind box unlocking request to the benefit platform server, receiving and displaying the benefits corresponding to the benefit blind box.

[0007] In a third aspect, embodiments of this application provide a rights blind box generation and sending device, applied to a rights platform server, comprising: The rights allocation module is used to dynamically cluster and divide rights into multiple rights sub-pools based on the attribute vectors of rights in the rights pool. The attribute vectors are ordered sequences formed by quantified values ​​corresponding to one or more predefined rights attributes arranged in a fixed order. The generation module is used to generate rights blind boxes in a hierarchical manner in response to the blind box acquisition command. The rights information in the rights blind boxes is encrypted and accompanied by a zero-knowledge proof to prove that the rights information meets predetermined conditions. The hierarchical manner is to select rights from each rights sub-pool in descending order of weight until the rights blind box is generated. The sending module is used to verify the validity of the zero-knowledge proof in response to the blind box unlocking command, and send the rights blind box to the user terminal device after successful verification.

[0008] In a fourth aspect, embodiments of this application provide a device for generating and sending blind boxes of rights, applied to a user terminal device, comprising: The first receiving module is used to send a rights blind box acquisition request to the rights platform server; the second receiving module is used to receive a rights blind box from the rights platform server, wherein the rights blind box is obtained based on the rights blind box generation and sending method described in the first aspect; the third receiving module is used to receive and display the rights corresponding to the rights blind box after sending a rights blind box unlocking request to the rights platform server.

[0009] In a fifth aspect, an electronic device is provided, comprising: a processor; a memory for storing executable instructions of the processor; wherein the processor is configured to execute the instructions to implement the method described in the first aspect above, and / or to implement the method described in the second aspect above.

[0010] In this embodiment, based on the attribute vectors of the rights in the rights pool, rights are dynamically clustered and divided into multiple rights sub-pools. In response to a blind box acquisition command, a hierarchical progressive method is used to generate rights blind boxes. The rights information in each rights blind box is encrypted and accompanied by a zero-knowledge proof proving that the rights information meets predetermined conditions. The hierarchical progressive method involves selecting rights from each rights sub-pool sequentially from highest to lowest weight until a rights blind box is generated. In response to a blind box unlock command, the validity of the zero-knowledge proof is verified, and after successful verification, the rights blind box is sent to the user's terminal device. That is, after obtaining multiple rights sub-pools, a hierarchical progressive technique and zero-knowledge proof technique are used to assist the rights center server in generating rights blind boxes. This method of generating rights blind boxes enhances the user's sense of surprise and experience, and improves the fairness and transparency of rights distribution. Users can receive and share exclusive blind boxes without revealing their VIP rights level, stimulating more users' desire to explore. Attached Figure Description

[0011] Figure 1 A flowchart illustrating a method for generating and sending a rights blind box, provided as an embodiment of this application; Figure 2 A flowchart illustrating another method for generating and sending benefit blind boxes provided in this application embodiment; Figure 3 A flowchart illustrating yet another method for generating and sending benefit blind boxes provided in this application embodiment; Figure 4 A flowchart illustrating yet another method for generating and sending benefit blind boxes provided in this application embodiment; Figure 5 A flowchart illustrating yet another method for generating and sending benefit blind boxes provided in this application embodiment; Figure 6 A flowchart illustrating yet another method for generating and sending benefit blind boxes provided in this application embodiment; Figure 7 A structural block diagram of a rights blind box generation and sending device provided in an embodiment of this application; Figure 8 A structural block diagram of another rights blind box generation and sending device provided in the embodiments of this application; Figure 9 This is a structural block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0012] The present application / disclosure will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application / disclosure and are not intended to limit the scope of the present application / disclosure. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present application / disclosure are shown in the accompanying drawings, not the entire structure.

[0013] The following description, in conjunction with the accompanying drawings, details a method for generating and sending benefit blind boxes provided in this application through specific embodiments and application scenarios.

[0014] Figure 1 This is a flowchart illustrating a method for generating and sending a rights blind box, provided in an embodiment of this application. The method can be executed by an electronic device, which may include a server and / or a terminal device. In other words, the method can be executed by software or hardware installed on the electronic device, and the method includes the following steps: Step 101: Based on the attribute vectors of the equity in the equity pool, the equity is dynamically clustered and divided into multiple equity sub-pools.

[0015] In some embodiments of this application, the rights platform records information such as the type, quantity, cost, and issuance probability of each type of rights into the rights platform according to operational needs, and creates a standardized VIP rights pool (i.e., rights pool). Based on the attribute vectors of the rights in the rights pool, the rights are dynamically clustered and divided to form multiple rights sub-pools. The attribute vector is an ordered sequence formed by quantified values ​​corresponding to one or more predefined rights attributes arranged in a fixed order.

[0016] It should be noted that the creation of the equity pool also includes the management of the equity pool, including configuration management, such as adjusting the issuance probability, adding or deleting equity types, in order to dynamically optimize the attractiveness and ROI (Return on Investment) of the equity pool.

[0017] Step 102: In response to the blind box acquisition instruction, a rights blind box is generated in a hierarchical manner. The rights information in the rights blind box is encrypted and accompanied by a zero-knowledge proof to prove that the rights information meets predetermined conditions. The hierarchical manner is to select rights from each rights sub-pool in descending order of weight until a rights blind box is generated.

[0018] Step 103: In response to the blind box unlocking command, verify the validity of the zero-knowledge proof, and after successful verification, send the rights blind box to the user terminal device.

[0019] In some embodiments of this application, in response to a blind box unlocking command, the validity of the zero-knowledge proof is verified, and upon successful verification, the benefit blind box is sent to the user's terminal device. The purpose of verifying the validity of the zero-knowledge proof is to ensure that the generation process of the benefit blind box is honest and legal. Anyone can verify the zero-knowledge proof to confirm that a benefit blind box meets predetermined conditions, but cannot know its specific contents.

[0020] Next, we will provide a detailed introduction to the implementation process of dynamically clustering and dividing the equity into multiple equity sub-pools based on the attribute vectors of the equity in the equity pool.

[0021] Figure 2 A flowchart illustrating another method for generating and sending benefit blind boxes provided in this application embodiment. (See attached flowchart.) Figure 2 As shown, based on the above embodiments, Figure 1 The implementation process of step 101 may include: Step 201: Receive and store the attribute information of each right to form a right pool. The attribute information includes at least one or more of the following: right type, right quantity, right cost, and right issuance probability.

[0022] Step 202: Extract the attribute vector of each equity in the equity pool. The attribute vector is obtained based on the attribute information of the equity.

[0023] Step 203: Based on the multi-dimensional weighted clustering algorithm, the rights in the rights pool are dynamically clustered and divided according to the attribute vector of each rights to form multiple rights sub-pools.

[0024] As one implementation method, the rights platform creates rights pools based on operational needs. Next, the platform utilizes a multi-dimensional weighted clustering algorithm, based on the attribute vectors of rights and interests. Clustering of rights and interests yields A dynamic equity sub-pool Each equity sub-pool Corresponding to a cluster center vector and a weighting coefficient Cluster center vector and weighting coefficients Satisfy the following formula (1):

[0025] in, Indicates the first The number of equity stakes in each equity sub-pool is given by the formula, where cj represents the cluster center vector and wj represents the weight coefficient. Multiple equity sub-pools can be obtained based on this formula. This adaptive equity sub-pooling mechanism can dynamically adjust according to actual conditions, making equity allocation more reasonable and efficient.

[0026] Figure 3 This is a flowchart illustrating yet another method for generating and sending benefit blind boxes, provided as an embodiment of this application. Figure 3 As shown, based on the above embodiments, Figure 1 The implementation process of step 102 may include: Step 301: In response to the blind box acquisition instruction, select a certain number of rights from each rights sub-pool in descending order of their weights and place them into the rights blind box to be generated.

[0027] Step 302: When the number of rights in the rights blind box to be generated reaches the preset number of rights, the rights blind box is obtained; When selecting a certain number of equity interests from the current equity sub-pool, the probability of those interests being selected is determined by the weight coefficient of the current equity sub-pool and an adjustable parameter.

[0028] In some embodiments of this application, the rights platform sets the total number of blind boxes. And the number of rights for each blind box Then, following a hierarchical order from high to low, rights are selected layer by layer from the rights sub-pool and placed into the rights blind box. Specifically, for the first... A blind box Its generation process is as follows: 1) From the equity sub-pool with the highest weight Random selection Individual rights included ,in The probability distribution is as follows:

[0029] in, express The number of benefits in the benefit sub-pool, where M represents the total number of benefits for each blind box. Represents a weight Positively correlated parameters, whose parameters control the transition from the first... The probability of selecting an equity from a sub-pool of equity.

[0030] Adjustable parameters The calculation formula is as follows:

[0031] in, This indicates an adjustable parameter that controls the degree to which the weights influence the selection probability. When hour, All sub-pools have an equal probability of being selected; when hour, It almost exclusively selects stakes from the first stake sub-pool. This is achieved through adjustment. A balance can be struck between "equality" and "efficiency".

[0032] 2) From the second-highest weighted equity sub-pool Random selection Individual rights included ,in The probability distribution is as follows:

[0033] in, express The number of benefits in the benefit sub-pool, where M represents the total number of benefits for each blind box. Represents a weight Positively correlated parameters, whose parameters control the transition from the first... The probability of selecting an equity from a sub-pool of equity.

[0034] 3) Continue in this manner until the sum of the number of rights selected from all rights sub-pools reaches [a certain value]. You will receive the final benefit blind box.

[0035] Based on the above-mentioned method for generating benefit blind boxes, this layered and progressive generation strategy can significantly enhance the richness and fun of blind box content.

[0036] In other embodiments of this application, to ensure the validity and legality of the blind box generation process, the rights information in the rights blind box is encrypted and accompanied by a zero-knowledge proof to prove that the rights information meets predetermined conditions. This includes: homomorphically encrypting the attribute vector of each right in the rights blind box to obtain an encrypted attribute vector; performing homomorphic operations on the encrypted attribute vectors of all rights in the rights blind box to generate an encrypted composite attribute vector, the encrypted composite attribute vector corresponding to the composite attribute of the rights blind box, the encrypted composite attribute vector encrypting the rights information; and generating a zero-knowledge proof based on the encrypted composite attribute vector, the zero-knowledge proof proving that the composite attribute corresponding to the encrypted composite attribute vector meets predetermined conditions, wherein the composite attribute of the rights blind box is constituted by the rights information in the rights blind box.

[0037] As an example, for the first A blind box Extract the attribute vector of its rights and interests. ,in Using homomorphic encryption, the encrypted attribute vector is obtained according to the following formula. :

[0038] in, It is a homomorphic encryption function (e.g., Paillier). Next, homomorphic operations are performed on the encrypted attribute vectors of all rights in the rights blind box to generate encrypted composite attribute vectors. The encrypted composite attribute vector Corresponding to the composite attributes of the aforementioned blind box of rights, the encrypted composite attribute vector To encrypt rights and interests information, The encrypted composite attribute vector is obtained according to the following formula. : Next, homomorphic operations are performed on these attribute vectors to obtain the composite attributes of the equity blind box. :

[0039] in, It is a composite attribute. It is the first The first right The weight coefficients of each attribute, This indicates the number of benefits contained in each benefit blind box. This indicates the number of attributes for each right. It is one of the public keys in homomorphic encryption, and consists of two large prime numbers. and The product of, i.e. . The bit length determines the security of the encryption; typically, 1024 bits or 2048 bits are chosen.

[0040] Finally, a zero-knowledge proof is generated based on the encrypted composite attribute vector. This zero-knowledge proof proves that the composite attribute corresponding to the encrypted composite attribute vector satisfies predetermined conditions, wherein the composite attribute of the rights and interests blind box is constituted by the rights and interests information in the blind box. In other words, the rights and interests platform uses zero-knowledge proof technology to prove the composite attribute without disclosing the specific rights and interests information of the blind box. Meets the predetermined conditions The formal representation of zero-knowledge proof is:

[0041] in, This represents zero-knowledge proof. Indicates encryption number The random number used when creating an attribute vector The symbol representing logical AND. Predefined condition. The design can be tailored to specific needs. For example, it can limit the rarity distribution and type combinations of benefits within the blind box, and predetermine conditions. This represents the range of values ​​for the composite attributes of the blind box, and is a predefined set. The setting of this range depends on the specific application scenario and requirements.

[0042] Generally speaking, pre-determined conditions It can be described by a set of inequalities or equations, for example:

[0043] in, and They represent the first The lower and upper bounds of the nth dimension. This range indicates that for the nth dimension... The first blind box, the first Composite attribute values ​​of each dimension Must be between and between.

[0044] Besides using inequalities, other mathematical tools can also be used to describe... Examples of suitable shapes include polygons or ellipsoids. The choice of form depends on the constraints imposed on the attributes of the blind box.

[0045] This method of verification protects user privacy while giving users a general understanding of the contents of the blind box, thus increasing their confidence in purchasing.

[0046] Figure 4 This is a flowchart illustrating yet another method for generating and sending benefit blind boxes, provided as an embodiment of this application. Figure 4 As shown, based on the above embodiments, Figure 1 The implementation process of step 103 may include: Step 401: In response to the blind box unlocking command, generate a verification triplet for the rights blind box. The verification triplet includes an encrypted attribute matrix, an encrypted composite attribute vector, and a zero-knowledge proof. The encrypted attribute matrix is ​​composed of the encrypted attribute vectors of all rights in the rights blind box. Step 402: Upload the verification triple to the blockchain network to verify the validity of the zero-knowledge proof; Step 403: After verification, the corresponding rights and benefits blind box is issued to the user's terminal device.

[0047] As one implementation method, the rights platform will use the encrypted attribute vector. Encrypted composite attribute vector and zero-knowledge proofs Uploaded to the blockchain, it undergoes verification by all nodes on the network. Data uploaded to the blockchain is represented as a triple. ,in: 1) Indicates the first A matrix consisting of the encrypted attribute vectors of all rights in each blind box:

[0048] in, Indicates the first The first blind box Each right's cryptographic attribute vector contains The encrypted value of each attribute.

[0049] 2) Indicates the first The encrypted composite attribute vector of a blind box is a Dimensional vector:

[0050] in, Indicates the first The first blind box The composite attribute value is obtained by analyzing the first [value] of all rights and interests. The result is obtained by weighted summation of the attributes.

[0051] 3) Indicates the first Zero-knowledge proofs for the blind box generation process are used to prove the encryption of composite attribute vectors. By analyzing the encrypted attribute matrix Obtained through legitimate operations, and possessing composite attributes Belongs to the predefined range .

[0052]

[0053] in, Indicates the generation of the encryption attribute matrix The random number matrix used at that time.

[0054] This triplet After being uploaded to the blockchain, all nodes on the network can verify it and check the zero-knowledge proof. The validity of the verification process ensures that the blind box production process is valid and legal. Anyone can verify this. To confirm the first The blind box meets the pre-order conditions, but its contents are unknown.

[0055] After a user purchases and opens a benefits blind box, the benefits platform extracts benefits from the corresponding benefits sub-pool based on the blind box's contents and distributes them to the user. Because the sub-pool is dynamically adjusted, the distribution of benefits is more equitable and reasonable.

[0056] Figure 5This is a flowchart illustrating yet another method for generating and sending benefit blind boxes, provided as an embodiment of this application. Figure 5 As shown, based on the above embodiments, the method further includes the following steps: Step 501: Collect behavioral data of target users during the acquisition of benefit blind boxes. The behavioral data includes the frequency, amount, and time of target users' purchase of benefit blind boxes, redemption data of benefit blind box prizes, sharing data of benefit blind boxes, and / or transfer data of benefit blind boxes.

[0057] Step 502: The collected benefit blind boxes are labeled and coded with attributes to extract feature information of the benefit blind boxes.

[0058] Step 503: Based on the sliding time window, calculate the frequency and amount of the target user obtaining the benefit blind box in each window.

[0059] Step 504: Based on the target user's behavioral data during the acquisition of the benefit blind box, the feature information of the benefit blind box, and the frequency and amount of the target user acquiring the benefit blind box in each window, feature engineering is constructed to determine whether the target user has any abnormal behavior.

[0060] In other words, this step targets and prevents fraudulent activities such as coupon discounts and promotions. Understandably, current anti-coupon-exploitation mechanisms are insufficient and struggle to effectively curb such behavior. The detection and prevention method proposed in this application, however, utilizes big data analysis and artificial intelligence to promptly identify abnormal transaction patterns and recognize fraudulent activities such as maliciously obtaining discounts and manipulating sales figures, thus protecting the legitimate rights and interests of platforms and merchants and creating a fair and trustworthy consumer environment.

[0061] In an exemplary embodiment, the data on the benefit blind boxes is first preprocessed, including: 1) collecting user behavior data during the blind box activity, the behavior data including the frequency, amount, and time of target users' purchases of benefit blind boxes, redemption data of benefit blind box prizes, sharing data of benefit blind boxes, and / or transfer data of benefit blind boxes; 2) labeling and encoding the benefit blind boxes with attributes. Key attributes of the blind boxes (such as brand, category, rarity) are extracted and converted into numerical or categorical features; 3) introducing time window features for the benefit blind boxes. Based on the duration of the blind box activity, different sales time windows are divided (such as the first hour, the first day, the first week), and user purchase behavior within each window is statistically analyzed. Secondly, after data preprocessing, based on the target users' behavior data during the acquisition of benefit blind boxes, the feature information of the benefit blind boxes, and the frequency and amount of benefit blind boxes acquired by the target users within each window, feature engineering is constructed to determine whether the target users exhibit abnormal behavior.

[0062] The following is an illustrative explanation of the feature engineering constructed based on the target user's behavioral data during the acquisition of benefit blind boxes, the feature information of benefit blind boxes, and the frequency and amount of benefit blind boxes acquired by the target user in each window. The constructed feature engineering includes a user-blind box preference matrix, a time pattern feature of purchasing benefit blind boxes, and a social influence feature, specifically including the following steps: 1) Construct a user-blind box preference matrix. Based on the target users' historical purchase records, calculate the target users' preference for the benefit attributes in different benefit blind boxes. :

[0063] in, Indicates user Blind box attributes preference Indicates user A collection of blind boxes I've purchased. Blind boxes representing targets In attributes Weight on, Indicates user blind boxes User ratings (such as purchase amount and frequency).

[0064] 2) Explore the time-based patterns of user blind box purchases. Statistically analyze the frequency and amount spent by target users on blind boxes within different time windows, and calculate their mean and variance:

[0065] in, and target users respectively In the time window Mean and variance of the frequency of in-app purchases of blind boxes. For time window The set of time steps (e.g., hours, days) within a given time period. For users At time step Number of times blind boxes were purchased in-app.

[0066] 3) Considering the social impact of users' blind box purchases, count the number of blind box shares and registrared items received by target users from their social networks (such as friends and group chats) as a social feature of their purchasing behavior. :

[0067] in, For users social networks and users respectively To users The number of times blind boxes are shared and given away.

[0068] As an example, based on the user-blind box preference matrix and the time pattern characteristics of target users' purchase of benefit blind boxes, the abnormal rating value of a target user's single purchase of benefit blind boxes is calculated.

[0069] Specifically, based on the user-blind box preference matrix and combined with the characteristics of user purchase time patterns, the "anomaly" of user purchase behavior is calculated. Purchase behaviors that deviate significantly from the average user purchase frequency and amount are given higher anomaly scores. :

[0070] in, Indicates the abnormal score value and users respectively No. Frequency and amount of each purchase and For users Historical average purchase frequency and amount, Indicates user No. The blind box purchased this time has attributes. The value that can be taken on. and This represents the weighting coefficient for time-based patterns, and its value can be obtained from real-time data analysis. Based on this formula, the abnormal rating values ​​for a single purchase of a benefit blind box by a target user can be obtained.

[0071] As an example, when calculating abnormal rating values, social influence features are introduced to correct the abnormal rating values, including: calculating the similarity of the target user's purchasing behavior with their social network; calculating a weight adjustment coefficient based on the purchasing behavior similarity; and obtaining the corrected abnormal rating value based on the purchasing behavior similarity and the weight adjustment coefficient.

[0072] Specifically, social influence features are incorporated into the weighted calculation of anomaly scores. Users with a high degree of similarity to purchasing behavior on social networks are given a higher anomaly weight. The outlier weights are obtained according to the following formula. :

[0073] in, Indicates user No. Abnormal ratings for subsequent purchases, i.e. , Indicates user No. The abnormal rating value of the purchase behavior, i.e. . For users Its social networks Similarity in purchasing behavior The weighting adjustment coefficients for social influence features. and Representing users respectively and its social networks Purchase behavior vector;

[0074] in, Indicates user In the The number of times you buy blind boxes express The average value of the purchasing behavior vectors of social network members. Purchasing behavior similarity is calculated using the following formula:

[0075] in, express Social network members in The average number of purchases on various blind boxes, with a similarity value ranging from... Within the range, a larger value indicates a higher user... The more similar the purchase behavior is to that of the individual on their social network, the more accurate the anomaly score becomes. This application obtains a corrected anomaly score based on the purchase behavior similarity and a weighting adjustment coefficient.

[0076] The following is an illustrative explanation of how to determine whether a target user exhibits abnormal behavior, which may include the following steps: Step 601: Assign different business importance weights to different types of benefit blind boxes; Step 602: Under the conditions of specific rights blind box attributes and time windows, based on the business importance weight, an anomaly judgment threshold is determined through an optimization process. Step 603: Determine whether the abnormal score value or the corrected abnormal score value is greater than the abnormal judgment threshold.

[0077] If the value is greater than 1, it proves that the target user is exhibiting abnormal behavior.

[0078] If it is less than, it proves that the target user does not have abnormal behavior.

[0079] As one implementation method, a hierarchical anomaly detection rule is established based on factors such as blind box attributes, purchase time, and social influence characteristics. Different detection thresholds are set for different time windows and social network structures.

[0080] in, Represents attributes In the time window The anomaly detection threshold within the range, This indicates taking the expression within the square brackets that reaches its maximum value. value. This represents the anomaly detection threshold, specifically, the anomaly score value greater than [a certain threshold]. The user was identified as an abnormal user due to unusual behavior. Indicates the first The importance weight of each type of blind box business should be considered. Different blind boxes may have different values, popularity, and risk levels, so they need to be treated differently when determining the threshold for anomaly detection. The value of is in Within the range, the larger the value, the more likely it is to be the first... The more important a type of blind box is, the greater its weight should be given in the anomaly detection.

[0081] This indicates that at the threshold value... The time window is Under these conditions, the probability that a user's behavior will be judged as abnormal can be estimated by analyzing the judgment results of abnormal users in historical data.

[0082] This indicates that at the threshold value... The time window is Under these conditions, the probability that a user's behavior is judged as normal can be estimated. Similarly, it can be estimated using the judgment results of normal users in historical data.

[0083] This step can determine whether a target user exhibits abnormal behavior. For target users exhibiting abnormal behavior, restrictions on blind box purchases or other control measures can be implemented. As one embodiment of this application, for users determined to have abnormal behavior, their abnormality score or a corrected abnormality score is compared with a set of tiered handling thresholds, and different levels of processing actions are executed based on the comparison results.

[0084] The following is another illustrative explanation of how to determine whether a target user exhibits abnormal behavior, which may include the following steps: Step 701: Obtain the target user's basic anomaly scores across different behavioral dimensions.

[0085] Step 702: Obtain the time dimension anomaly score and the social dimension anomaly score.

[0086] Step 703: The multiple basic anomaly scores, time-dimensional anomaly scores, and social-dimensional anomaly scores are weighted and summed according to preset weights to generate a comprehensive anomaly index.

[0087] As one implementation method, multi-dimensional anomaly scoring is integrated to generate a comprehensive anomaly index for users. In addition to considering factors such as preference anomalies and group similarity, the scoring also takes into account factors such as temporal regularity and social influence.

[0088] in, Indicates comprehensive abnormal indicators, Indicates the first The weighting coefficients for each dimension of the anomaly rating. Indicates user In the Basic anomaly scores across all dimensions. and Representing users respectively Anomaly measures in terms of temporal patterns and social influence, including: 1) Time dimension anomaly scoring : First, the user The purchasing behavior is arranged in chronological order to obtain a time series. ,in express The timestamp of the first purchase. Indicates the first The amount or quantity of the purchase.

[0089] Next, the Local Outlier Factor (LOF) at each time point is calculated to measure the degree of anomalousness of that point relative to its neighborhood, thereby yielding... .

[0090]

[0091] in, express of Nearest time point set, express The locally achievable density. Indicates a point in time and Distance metric between.

[0092] 2) Abnormal scores in the social dimension : First, build users social networks For a set of user nodes, This is a set of edges representing relationships between users.

[0093] Next, calculate the user The influence score measures its abnormal propagation ability, and ultimately yields... .

[0094]

[0095] in, Indicates user For users The influence This is the influence threshold parameter. Indicates user and similarity, and These represent the mean and standard deviation of the influence scores for all users, respectively.

[0096] After generating the comprehensive anomaly index, the method further includes: presetting multiple sequentially increasing handling level thresholds; comparing the comprehensive anomaly index with the multiple handling level thresholds to determine the numerical range in which the comprehensive anomaly index is located; and automatically executing the handling action corresponding to the level of the determined numerical range.

[0097] By introducing a sliding time window mechanism and a multi-dimensional anomaly scoring mechanism, real-time, comprehensive, and robust detection of user "coupon-clipping" behavior is achieved. This algorithm can quickly capture dynamically changing anomaly patterns, output interpretable anomaly scores, and flexibly adapt to new coupon-clipping techniques. It has made fruitful progress in the field of anti-coupon-clipping, forming a practical and efficient anomaly detection scheme.

[0098] Specifically, users who engage in "coupon hunting" (abnormal users) will be dealt with in a tiered manner, with levels of action set according to the severity of the user's abnormality, such as "warning - purchase limit - freeze - ban".

[0099] in, These are thresholds for handling different levels of fraudulent activities. By implementing tiered handling, we can curb such activities while minimizing the impact on legitimate users.

[0100] As one embodiment of this application, the anomaly detection threshold is dynamically adjusted to adapt to the sales rhythm of the blind box activity. Based on the distribution of user behavior at different stages of the blind box activity (e.g., pre-sale period, peak period, and post-sale period), the anomaly detection threshold is adaptively adjusted.

[0101] in, The adjusted anomaly detection threshold. As the initial threshold, For decay rate, This is the start time of the activity. As the activity progresses, the anomaly detection threshold is gradually lowered to adapt to changes in user behavior.

[0102] The human-computer interaction process for generating blind box anomaly detection is used to understand the actual motivation and scenario of users who purchase blind boxes and to collect user feedback.

[0103]

[0104] Based on the verification results, the algorithm's anomaly detection was corrected and optimized to improve its business adaptability.

[0105] Next, we will introduce the implementation process of the method for generating and sending benefit blind boxes applied to user terminal devices. Figure 6 This is a flowchart illustrating yet another method for generating and sending benefit blind boxes, provided as an embodiment of this application. It is applied to user terminal devices. Figure 6 As shown, the method may include the following steps: Step 801: Send a request to the rights platform server to obtain the rights blind box.

[0106] Step 802: Receive an equity blind box from the equity platform server, wherein the equity blind box is generated by the equity platform server by executing the equity blind box generation and sending method. Step 803: After sending a rights blind box unlocking request to the rights platform server, receive and display the rights corresponding to the rights blind box.

[0107] As one embodiment of this application, before receiving and displaying the rights corresponding to the rights blind box, a verification status information is also displayed. The verification status information is used to prove that the verification triplet corresponding to the rights blind box has been uploaded to the blockchain network, and the zero-knowledge proof therein has passed the validity verification.

[0108] By innovatively combining zero-knowledge proof technology with privilege blind boxes, a brand-new interactive approach to marketing campaigns is provided. Users can claim and share exclusive privilege blind boxes without revealing their VIP privilege level, sparking more users' desire to explore and driving high-frequency dissemination and interactive conversion of marketing content. This novel marketing model helps brands stand out in a highly competitive market and attract more potential users.

[0109] The appeal of blind boxes can effectively drive high-value consumer behavior, directly contributing to revenue growth. Simultaneously, the scarcity and social sharing attributes of blind boxes can accelerate the viral spread of brand influence, attracting more new users, expanding the brand's potential customer base, and providing a continuous stream of new momentum for business growth.

[0110] As one embodiment of this application, the method for generating and sending benefit blind boxes will be further explained in conjunction with a specific application scenario.

[0111] Step 1: Creation and Management of the VIP Benefits Pool: Step Two: Implementation of the Zero-Knowledge Proof-Based Stake Blind Box Generation Protocol: This application introduces zero-knowledge proof technology into the stake blind box generation process, achieving efficient and reliable stake verification and blind box generation while protecting user privacy. The algorithm supports flexible attribute proofs, possesses strong versatility and scalability, and balances privacy security, trust guarantees, and system performance.

[0112] Step 3: Box Distribution and Sales 1) The rights center operators promote the rights blind boxes to target users through various marketing channels and activities, such as APP push, SMS, and advertisements; 2) Users purchase blind boxes through the benefits platform. The platform then calls the blind box generation module to randomly generate blind boxes and distribute them to users. Simultaneously, the platform records key information such as the blind box's generation parameters and encrypted content, storing this information in a database for subsequent verification.

[0113] Step Four: Unlocking the Blind Box and Distributing Benefits 1) When target users unlock blind boxes, the rights platform verifies the validity of the rights blind boxes, publishes the contents of the rights blind boxes, issues corresponding VIP rights to users, and triggers corresponding marketing actions, such as APP notifications or SMS messages. 2) Link the VIP benefits obtained by users to their accounts and synchronize them to the corresponding business system for users to use; Step 5: Anti-fraud mechanism based on sliding time window: Restrict blind box purchases or other control measures for abnormal users; Step Six: Building a Machine Learning-Driven Risk Control System: Constructing an intelligent risk control system for rights and interests. The engine continuously optimizes the risk control model by mining multi-dimensional data such as user behavior, transaction records, coupon-hunting records, or device fingerprints.

[0114] Step 7: Tracking and Analysis of Rights and Benefits Usage: By collecting logs and cleaning data, a behavioral data warehouse for the rights and benefits usage chain is built, and a rights and benefits usage dashboard is designed. Operations personnel can use the dashboard to understand the usage of rights and benefits, including key indicators such as usage frequency, average spending per person, and conversion rate.

[0115] Step 8: Cultivating an On-Chain Reputation System: Utilizing the ledger characteristics of blockchain, an immutable on-chain reputation profile is established for each user. Based on various user behaviors on the platform, such as purchases, cheating, and complaints, the user's reputation score is updated in real time. The reputation score is linked to user rights to incentivize active participation and positive behavior, while taking appropriate action against users who breach trust.

[0116] To implement the above embodiments, this application provides a device for generating and sending blind boxes of rights.

[0117] Figure 7 This is a structural block diagram of a rights blind box generation and sending device provided in an embodiment of this application. The rights blind box generation and sending device of this embodiment is applied to a rights platform server. Figure 7 As shown, the device includes: The rights and interests partitioning module 901 is used to dynamically cluster and partition the rights and interests based on the attribute vectors of the rights and interests in the rights and interests pool to form multiple rights and interests sub-pools. The attribute vectors are ordered sequences formed by quantified values ​​corresponding to one or more predefined rights and interests attributes arranged in a fixed order. The generation module 902 is used to generate a rights blind box in response to the blind box acquisition instruction, using a hierarchical and progressive method. The rights information in the rights blind box is encrypted and accompanied by a zero-knowledge proof to prove that the rights information meets predetermined conditions. The hierarchical and progressive method is to select rights from each rights sub-pool in descending order of weight until a rights blind box is generated. The sending module 903 is used to respond to the blind box unlocking command, verify the validity of the zero-knowledge proof, and send the rights blind box to the user terminal device after the verification is successful.

[0118] In some embodiments of this application, the rights allocation module 901 is specifically used for: Receive and store the attribute information of each right to form a right pool. The attribute information includes at least one or more of the following: right type, right quantity, right cost, and right issuance probability. Extract the attribute vector of each equity in the equity pool, wherein the attribute vector is obtained based on the attribute information of the equity; Based on a multi-dimensional weighted clustering algorithm, the rights in the rights pool are dynamically clustered and divided according to the attribute vector of each rights, forming multiple rights sub-pools.

[0119] In some embodiments of this application, the generation module 902 is specifically used for: In response to the blind box acquisition command, a certain number of rights are selected from each rights sub-pool in descending order of weight and placed into the rights blind box to be generated. When the number of rights in the rights blind box to be generated reaches the preset number of rights, the rights blind box is obtained; When selecting a certain number of equity interests from the current equity sub-pool, the probability of those interests being selected is determined by the weight coefficient of the current equity sub-pool and an adjustable parameter.

[0120] In some embodiments of this application, the sending module 903 is specifically used for: In response to the blind box unlocking command, a verification triplet for the rights blind box is generated. The verification triplet includes an encrypted attribute matrix, an encrypted composite attribute vector, and a zero-knowledge proof. The encrypted attribute matrix is ​​composed of the encrypted attribute vectors of all rights in the rights blind box. The verification triple is uploaded to the blockchain network to verify the validity of the zero-knowledge proof; Once verified, the corresponding benefits blind box will be issued to the user's terminal device.

[0121] In some embodiments of this application, the device further includes: The collection module 1101 is used to collect behavioral data of target users during the process of acquiring benefit blind boxes. The behavioral data includes the frequency, amount and time of target users purchasing benefit blind boxes, redemption data of benefit blind box prizes, sharing data of benefit blind boxes and / or transfer data of benefit blind boxes. The extraction module 1102 is used to perform attribute labeling and encoding on the collected rights blind boxes in order to extract the feature information of the rights blind boxes; The statistics module 1103 is used to calculate the frequency and amount of the target user obtaining the benefit blind box in each window based on the sliding time window. The determination module 1103 is used to construct feature engineering based on the target user's behavioral data during the acquisition of the benefit blind box, the feature information of the benefit blind box, and the frequency and amount of the target user acquiring the benefit blind box in each window, so as to determine whether the target user has abnormal behavior.

[0122] As one possible implementation, module 1103 is specifically used for: Based on the weights of the blind box attributes in the target user's historical purchase records and the user ratings, a user-blind box preference matrix is ​​constructed. The user-blind box preference matrix is ​​used to calculate the target user's preference for the blind box attributes in the blind boxes. The user ratings are obtained based on the behavioral data and the feature information of the blind boxes. Based on the frequency and amount of target users purchasing benefit blind boxes in different time windows, we can obtain the time pattern characteristics of target users purchasing benefit blind boxes. The data on the sharing and transfer of benefit blind boxes received by target users from social networks are statistically analyzed as social influence characteristics; The constructed feature engineering includes the constructed user-blind box preference matrix, the constructed time pattern features of purchasing rights blind boxes, and the constructed social influence features.

[0123] The determination module 1103 is also used for: Assign different business importance weights to different types of benefit blind boxes; Under specific rights and interests blind box attributes and time windows, an anomaly detection threshold is determined through an optimization process based on the aforementioned business importance weight. Determine whether the abnormal score value or the corrected abnormal score value is greater than the abnormal judgment threshold; If the value is greater than 1, it proves that the target user has abnormal behavior. If it is less than, it proves that the target user does not have abnormal behavior.

[0124] Figure 8 This is a structural block diagram of another rights blind box generation and sending apparatus provided in an embodiment of this application. It should be noted that this apparatus is applied to a user terminal device. Figure 8 As shown, the device includes: The acquisition module 1201 is used to send a rights blind box acquisition request to the rights platform server; The first receiving module 1202 is used to receive an equity blind box from the equity platform server, wherein the equity blind box is generated by the equity platform server by executing the equity blind box generation and sending method; The second receiving module 1203 is used to receive and display the rights corresponding to the rights blind box after sending a rights blind box unlocking request to the rights platform server.

[0125] In some embodiments of this application, the device further includes: The verification module 1301 is used to display verification status information, which proves that the verification triple corresponding to the equity blind box has been uploaded to the blockchain network and that the zero-knowledge proof therein has passed the validity verification.

[0126] Optional, such as Figure 9 As shown, this application embodiment also provides an electronic device 60, including a processor 601, a memory 602, and a program or instructions stored in the memory 602 and executable on the processor 601. When the program or instructions are executed by the processor 601, they implement the various processes of the above-described rights blind box generation and sending method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0127] This application also provides a computer-readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described method embodiments and achieve the same technical effects. To avoid repetition, these will not be described again here.

[0128] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0129] This application provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, which, when executed by a computer, cause the computer to perform the various processes described in the above method embodiments and achieve the same technical effect. To avoid repetition, these will not be repeated here.

[0130] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0131] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0132] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for generating and sending benefit blind boxes, applied to a benefit platform server, characterized in that, include: Based on the attribute vectors of the rights in the rights pool, the rights are dynamically clustered and divided into multiple rights sub-pools. The attribute vectors are ordered sequences formed by quantified values ​​corresponding to one or more predefined rights attributes arranged in a fixed order. In response to the blind box acquisition command, a hierarchical and progressive approach is adopted to generate a rights blind box. The rights information in the rights blind box is encrypted and accompanied by a zero-knowledge proof to prove that the rights information meets the predetermined conditions. The hierarchical and progressive approach is to select rights from each rights sub-pool in descending order of weight until a rights blind box is generated. In response to the blind box unlocking command, the validity of the zero-knowledge proof is verified, and after successful verification, the rights blind box is sent to the user's terminal device.

2. The method according to claim 1, characterized in that, Based on the attribute vectors of the equity in the equity pool, the equity is dynamically clustered and divided into multiple equity sub-pools, including: Receive and store the attribute information of each right to form a right pool. The attribute information includes at least one or more of the following: right type, right quantity, right cost, and right issuance probability. Extract the attribute vector of each equity in the equity pool, wherein the attribute vector is obtained based on the attribute information of the equity; Based on a multi-dimensional weighted clustering algorithm, the rights in the rights pool are dynamically clustered and divided according to the attribute vector of each rights, forming multiple rights sub-pools.

3. The method according to claim 1, characterized in that, In response to the blind box acquisition command, a layered and progressive approach is adopted to select benefits from the benefit sub-pool layer by layer to generate benefit blind boxes, including: In response to the blind box acquisition command, a certain number of rights are selected from each rights sub-pool in descending order of weight and placed into the rights blind box to be generated. When the number of rights in the rights blind box to be generated reaches the preset number of rights, the rights blind box is obtained; When selecting a certain number of equity interests from the current equity sub-pool, the probability of those interests being selected is determined by the weight coefficient of the current equity sub-pool and an adjustable parameter.

4. The method according to claim 1, characterized in that, The rights information in the rights blind box is encrypted and accompanied by a zero-knowledge proof to demonstrate that the rights information meets predetermined conditions, including: Homomorphic encryption is performed on the attribute vector of each right within the right blind box to obtain the encrypted attribute vector; Homomorphic operations are performed on the encrypted attribute vectors of all rights in the rights blind box to generate an encrypted composite attribute vector. The encrypted composite attribute vector corresponds to the composite attribute of the rights blind box, and the encrypted composite attribute vector realizes the encryption of rights information. A zero-knowledge proof is generated based on the encrypted composite attribute vector. The zero-knowledge proof is used to prove that the composite attribute corresponding to the encrypted composite attribute vector satisfies a predetermined condition, wherein the composite attribute of the rights and interests box is constituted based on the rights and interests information in the rights and interests box.

5. The method according to claim 4, characterized in that, In response to the blind box unlocking command, the validity of the zero-knowledge proof is verified, and upon successful verification, the rights blind box is sent to the user's terminal device, including: In response to the blind box unlocking command, a verification triplet for the rights blind box is generated. The verification triplet includes an encrypted attribute matrix, an encrypted composite attribute vector, and a zero-knowledge proof. The encrypted attribute matrix is ​​composed of the encrypted attribute vectors of all rights in the rights blind box. The verification triple is uploaded to the blockchain network to verify the validity of the zero-knowledge proof; Once verified, the corresponding benefits blind box will be issued to the user's terminal device.

6. The method according to claim 1, characterized in that, After sending the rights and benefits blind box to the user terminal device, the method further includes: Collect behavioral data of target users during the acquisition of benefit blind boxes. The behavioral data includes the frequency, amount and time of target users' purchase of benefit blind boxes, redemption data of benefit blind box prizes, sharing data of benefit blind boxes and / or transfer data of benefit blind boxes. The collected benefit blind boxes are labeled and coded with attributes to extract their feature information; Based on the sliding time window, the frequency and amount of the target user obtaining the benefit blind box in each window were statistically analyzed; Based on the target user's behavioral data during the acquisition of benefit blind boxes, the feature information of benefit blind boxes, and the frequency and amount of benefit blind boxes acquired by the target user in each window, feature engineering is constructed to determine whether the target user has any abnormal behavior.

7. The method according to claim 6, characterized in that, Based on the target user's behavioral data during the acquisition of benefit blind boxes, the feature information of the benefit blind boxes, and the frequency and amount of benefit blind boxes acquired by the target user in each window, feature engineering is constructed, including: Based on the weights of the blind box attributes in the target user's historical purchase records and the user ratings, a user-blind box preference matrix is ​​constructed. The user-blind box preference matrix is ​​used to calculate the target user's preference for the blind box attributes in the blind boxes. The user ratings are obtained based on the behavioral data and the feature information of the blind boxes. Based on the frequency and amount of target users purchasing benefit blind boxes in different time windows, we can obtain the time pattern characteristics of target users purchasing benefit blind boxes. The data on the sharing and transfer of benefit blind boxes received by target users from social networks are statistically analyzed as social influence characteristics; The constructed feature engineering includes the constructed user-blind box preference matrix, the constructed time pattern features of purchasing rights blind boxes, and the constructed social influence features.

8. The method according to claim 7, characterized in that, The method further includes: Based on the user-blind box preference matrix and the time pattern characteristics of target users' purchase of benefit blind boxes, the abnormal rating values ​​of target users when purchasing benefit blind boxes in a single transaction are calculated.

9. The method according to claim 8, characterized in that, When calculating abnormal rating values, social influence features are introduced to correct the abnormal rating values, specifically including: Calculate the similarity between the target user's purchasing behavior and that of their social network; Calculate the weight adjustment coefficient based on the similarity of the purchasing behaviors; Based on the similarity of the purchase behavior and the weight adjustment coefficient, the corrected abnormal score value is obtained.

10. The method according to claim 9, characterized in that, Determine whether the target user exhibits any unusual behavior, including: Assign different business importance weights to different types of benefit blind boxes; Under specific rights and interests blind box attributes and time windows, an anomaly detection threshold is determined through an optimization process based on the aforementioned business importance weight. Determine whether the abnormal score value or the corrected abnormal score value is greater than the abnormal judgment threshold; If the value is greater than 1, it proves that the target user has abnormal behavior. If it is less than, it proves that the target user does not have abnormal behavior.

11. The method according to claim 9, characterized in that, The method further includes: Obtain multiple basic anomaly scores for the target user across different behavioral dimensions; Obtain anomaly scores in the time dimension and the social dimension; The multiple basic anomaly scores, time-dimensional anomaly scores, and social-dimensional anomaly scores are weighted and summed according to preset weights to generate a comprehensive anomaly index.

12. The method according to claim 11, characterized in that, After generating the comprehensive anomaly index, the following is also included: Multiple threshold values ​​for treatment levels are preset and increase sequentially. The comprehensive anomaly index is compared with the multiple treatment level thresholds to determine the numerical range of the comprehensive anomaly index. Based on the determined numerical range, the corresponding level of action will be automatically executed.

13. The method according to claim 10, characterized in that, The method further includes: For users identified as having abnormal behavior, their abnormal score or corrected abnormal score is compared with a set of graded handling thresholds, and different levels of handling actions are performed based on the comparison results.

14. The method according to claim 10, characterized in that, The anomaly detection threshold is adaptively adjusted according to different stages of the benefit blind box activity, and the anomaly detection threshold gradually decreases to adapt to changes in the target user's behavior.

15. A method for generating and sending benefit blind boxes, applied to user terminal equipment, characterized in that, include: Send a request to the rights and benefits platform server to obtain the rights and benefits blind box; A rights blind box is received from the rights platform server, wherein the rights blind box is generated by the rights platform server by performing the method as described in any one of claims 1 to 14; After sending a request to unlock the blind box to the rights platform server, the rights corresponding to the blind box are received and displayed.

16. The method according to claim 15, characterized in that, Before receiving and displaying the rights corresponding to the rights blind box, the process also includes: Display a verification status message, which proves that the verification triplet corresponding to the equity blind box has been uploaded to the blockchain network, and that the zero-knowledge proof therein has passed the validity verification.

17. A device for generating and sending benefit blind boxes, applied to a benefit platform server, characterized in that, include: The equity partitioning module is used to dynamically cluster and partition the equity based on the attribute vectors of the equity in the equity pool, forming multiple equity sub-pools. The attribute vector is an ordered sequence formed by quantified values ​​corresponding to one or more predefined equity attributes arranged in a fixed order. The generation module is used to generate a rights blind box in response to the blind box acquisition instruction, and to generate rights blind boxes in a hierarchical manner. The rights information in the rights blind box is encrypted and accompanied by a zero-knowledge proof to prove that the rights information meets the predetermined conditions. The hierarchical manner is to select rights from each rights sub-pool in descending order of weight until the rights blind box is generated. The sending module is used to respond to the blind box unlocking command, verify the validity of the zero-knowledge proof, and send the rights blind box to the user terminal device after the verification is successful.

18. A device for generating and sending blind boxes of benefits, applied to user terminal equipment, characterized in that, include: The acquisition module is used to send a request to the rights platform server to acquire the rights blind box; A first receiving module is configured to receive an equity blind box from the equity platform server, wherein the equity blind box is generated by the equity platform server by performing the method as described in any one of claims 1 to 14; The second receiving module is used to receive and display the rights corresponding to the rights blind box after sending a rights blind box unlocking request to the rights platform server.

19. An electronic device, characterized in that, include: processor; Memory for storing the executable instructions of the processor; The processor is configured to execute the instructions to implement the method as described in any one of claims 1-14, and / or to implement the method as described in any one of claims 15-16.

Images