Benefit resource allocation method and device, storage medium and program product

Abstract

One or more embodiments of the invention provide a right resource allocation method and device, a storage medium and a program product. The rights and interests configuration method comprises the steps of determining a target crowd to participate in a resource allocation activity; the interest sensitive score of each user in the target group is acquired from a pre-stored interest sensitive score data set, the interest sensitive score is calculated by a preset transaction prediction model, and the transaction prediction model comprises two output branches: one branch is used for outputting the transaction probability of the user under the condition that the interest is not issued, and the other branch is used for outputting the transaction probability of the user under the condition that the interest is not issued; the other branch is used for outputting the transaction probability of the user under the condition of issuing rights and interests, and the rights and interests sensitive score is used for representing the difference of the two transaction probabilities; according to the method, the target population is divided into at least two target sub-populations according to the right sensitive score of the user, and then the target right amount is configured for the at least two target sub-populations in a differentiated manner, so that more accurate resource release and distribution are realized.

Description

(19) State Intellectual Property Office (12) Invention Patent Application (10) Application Publication Number (43) Application Publication Date (21) Application Number 202511235415.1 (22) Application Date 2025.08.29 (71) Applicant Ant Blockchain Technology (Shanghai) Co., Ltd. Address Room 803, 8th Floor, No. 618 Waima Road, Huangpu District, Shanghai 200010 (72) Inventors Liu Wenyan, Li Weixian, Yin Shan (74) Patent Agency Beijing Bosijia Intellectual Property Agency Co., Ltd. 11415 Patent Attorney Chen Yurou (51) Int.Cl. G06Q 30 / 0207 (2023.01) G06Q 30 / 0202 (2023.01) (54) Invention Title: Method, Device, Storage Medium and Program Product for Allocating Rights and Resources (57) Abstract This specification provides a method, device, storage medium and program product for allocating rights and resources through one or more embodiments. The rights and benefits allocation method includes: identifying the target population to participate in the resource allocation activity; obtaining the rights and benefits sensitivity scores of each user in the target population from a pre-stored rights and benefits sensitivity score dataset. The rights and benefits sensitivity scores are calculated by a pre-set transaction prediction model, which includes two output branches: one branch outputs the transaction probability of a user when no rights and benefits are issued, and the other branch outputs the transaction probability of a user when rights and benefits are issued. The rights and benefits sensitivity scores are used to characterize the difference between these two transaction probabilities; dividing the target population into at least two target sub-groups based on the users' rights and benefits sensitivity scores, and then differentiating the target rights and benefits amounts for at least two target sub-groups to achieve more precise resource allocation and distribution. Claims (2 pages), Description (13 pages), Drawings (3 pages), CN 121169473 A, 2025.12.19, CN 1 21 16 94 73 A. 1. A method for allocating rights and interests resources, comprising: determining a target population to participate in a resource allocation activity; querying a pre-stored data set of rights and interests sensitive scores for users in the target population; the rights and interests sensitive scores of each user in the data set are predicted using a preset transaction prediction model; the transaction prediction model includes two output branches, one output branch outputting the transaction probability of a user when no rights and interests are distributed, and the other output branch outputting the transaction probability of a user when rights and interests are distributed, the rights and interests sensitive scores representing the difference between the transaction probability of a user when rights and interests are distributed and the transaction probability when no rights and interests are distributed; dividing the target population into at least two target sub-populations based on the rights and interests sensitive scores of users in the target population, such that the rights and interests sensitive scores of users in different target sub-populations belong to different ranges of rights and interests sensitive scores; and differentially configuring the at least two target sub-populations based on the respective ranges of rights and interests sensitive scores corresponding to the at least two target sub-populations.1. The target benefit amount corresponding to each target subgroup. 2. The method according to claim 1, wherein dividing the target population into at least two target subgroups based on the benefit sensitivity scores of users in the target population includes: sorting the benefit sensitivity scores of users in the target population in descending order to obtain a benefit sensitivity score curve; determining at least one target position point from the benefit sensitivity score curve, wherein the curvature of the target position point is greater than the curvature of at least some other position points in the benefit sensitivity score curve; obtaining at least two benefit sensitivity score ranges based on the benefit sensitivity scores corresponding to the at least one target position point; dividing the target population into at least two target subgroups based on the benefit sensitivity scores of users in the target population and the at least two benefit sensitivity score ranges. 3. The method according to claim 1, wherein the step of differentially configuring the target equity amount corresponding to the at least two target subgroups based on the equity sensitivity ranges corresponding to the at least two target subgroups includes: determining a baseline equity amount that satisfies the equity upper limit constraint set by the resource allocation activity; differentially adjusting the baseline equity amount based on the equity sensitivity ranges corresponding to the at least two target subgroups to obtain the target equity amounts corresponding to the at least two target subgroups respectively; if the target equity amounts corresponding to the at least two target subgroups do not satisfy the equity upper limit constraint, correcting the target equity amount of at least one of the target subgroups until the target equity amounts corresponding to the at least two target subgroups satisfy the equity upper limit constraint. 4. The method according to claim 3, wherein if the target equity amounts corresponding to the at least two target subgroups do not meet the upper limit constraint, the target equity amounts of at least one target subgroup are corrected until the target equity amounts corresponding to the at least two target subgroups meet the upper limit constraint, comprising: obtaining statistical weights for the target equity amounts corresponding to each target subgroup; performing weighted statistics based on the target equity amounts corresponding to the at least two target subgroups and the statistical weights to obtain weighted statistical equity amounts; if the weighted statistical equity amounts are greater than the upper limit constraint, correcting the target equity amounts of at least one target subgroup, and performing weighted statistics again until the weighted statistical equity amounts obtained again are not greater than the upper limit constraint. Claims 1 / 2 Page 2 CN 121169473 A 5. The method according to claim 4, wherein the statistical weight of the target equity corresponding to the target subgroup is determined by the estimated transaction probability of the target subgroup; or, the statistical weight of the target equity corresponding to the target subgroup is determined by the estimated number of transactions of the target subgroup, wherein the estimated number of transactions of the target subgroup includes the product between the estimated transaction probability of the target subgroup and the total number of the target subgroup;The estimated transaction probability of the target subgroup is determined as follows: Transaction information of historical participants in the resource allocation activity is obtained, including the rights-sensitive sub-segments to which users in the historical subgroups belong and the transaction probabilities corresponding to each rights-sensitive sub-segment; If the target subgroup corresponds to a single rights-sensitive sub-segment, the transaction probability corresponding to that rights-sensitive sub-segment is determined as the estimated transaction probability of the target subgroup; If the target subgroup involves at least two rights-sensitive sub-segments, the estimated transaction probability of the target subgroup is calculated based on the transaction probabilities corresponding to the at least two rights-sensitive sub-segments, combined with the proportion of the target subgroup in each sensitive sub-segment. 6. The method according to claim 3, wherein the step of differentially adjusting the benchmark equity amount based on the equity sensitivity score ranges corresponding to the at least two target subgroups to obtain the target equity amounts corresponding to the at least two target subgroups includes: determining the adjustment weight corresponding to each equity sensitivity score range, wherein the adjustment weight corresponding to each equity sensitivity score range is positively correlated with the equity sensitivity score in that equity sensitivity score range; and performing a weighted adjustment on the benchmark equity amount based on the adjustment weight corresponding to each equity sensitivity score range to obtain the target equity amount applicable to the target subgroups corresponding to that range. 7. The method according to claim 4, wherein adjusting the target equity amount of at least one target subgroup comprises: determining at least one target subgroup to be adjusted from the at least two target subgroups based on the equity sensitivity ranges corresponding to the at least two target subgroups and the preset lower limit of equity amount for each target subgroup; and reducing the target equity amount of the target subgroup to be adjusted to a value not lower than the preset lower limit of equity amount for the target subgroup to be adjusted; or, determining a normalization coefficient based on the ratio between the weighted statistical equity amount and the equity upper limit constraint, and normalizing the target equity amount corresponding to each target subgroup based on the normalization coefficient. 8. An electronic device comprising: a processor; a memory for storing processor-executable instructions; wherein the processor implements the steps of the method according to any one of claims 1 to 7 by executing the executable instructions. 9. A computer-readable storage medium storing computer instructions thereon, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7. 10. A computer program product comprising a computer program / instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 7. Claims 2 / 2 Page 3 CN 121169473 A Rights Resource Allocation Method, Apparatus, Storage Medium and Program Product Technical Field

[0001] One or more embodiments of this specification relate to the field of computer technology, and more particularly to a rights resource allocation method...Law, electronic device, computer-readable storage medium, and computer program product. Background Art

[0002] In today's era of booming digital economy, various commercial institutions and platforms often carry out various forms of resource allocation activities to attract users and promote transactions. Among them, the reasonable allocation of rights and interests is crucial.

[0003] At present, most institutions, after determining the target group to participate in resource allocation activities, often adopt a relatively simple and extensive rights and interests allocation method. For example, the common practice is to treat the target group equally and uniformly issue the same type and quantity of rights and interests, or to preliminarily classify them according to some basic user attributes, such as consumption amount range, registration time, etc., and then give each group a fixed rights and interests allocation plan, which results in low utilization of rights and interests resources. Summary of the Invention

[0004] In view of this, one or more embodiments of this specification provide a method for allocating rights and interests resources, an electronic device, a computer-readable storage medium, and a computer program product.

[0005] To achieve the above objectives, one or more embodiments of this specification provide the following technical solutions:

[0006] According to a first aspect of one or more embodiments of this specification, a method for allocating rights and interests resources is proposed, comprising:

[0007] determining a target population to participate in a resource allocation activity;

[0008] querying a pre-stored data set of rights and interests sensitive scores to obtain the rights and interests sensitive scores of users in the target population, wherein the rights and interests sensitive scores of each user in the data set of rights and interests sensitive scores are predicted using a preset transaction prediction model; the transaction prediction model includes two output branches, one output branch for outputting the transaction probability of a user when no rights and interests are distributed, and the other output branch for outputting the transaction probability of a user when rights and interests are distributed, wherein the rights and interests sensitive scores are used to characterize the difference between the transaction probability of a user when rights and interests are distributed and the transaction probability when no rights and interests are distributed;

[0009] dividing the target population into at least two target sub-populations based on the rights and interests sensitive scores of users in the target population, such that the rights and interests sensitive scores of users in different target sub-populations belong to different ranges of rights and interests sensitive scores;

[0010] Based on the rights-sensitive segment ranges corresponding to the at least two target subgroups, the target rights amounts corresponding to the at least two target subgroups are configured differently.

[0011] According to a second aspect of the embodiments of this specification, an electronic device is provided, comprising:

[0012] a processor;

[0013] a memory for storing processor-executable instructions;

[0014] wherein, when the processor executes the executable instructions, it is used to implement the method described in the first aspect.

[0015] According to a third aspect of the embodiments of this specification, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0016] According to a fourth aspect of the embodiments of this specification, a computer program product is provided, including a computer program, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0017] The technical solutions provided by the embodiments of this specification may include the following beneficial effects:

[0018] In the embodiments of this specification, after determining the target population to participate in the resource allocation activity, the corresponding equity sensitivity score can be directly matched from the pre-stored equity sensitivity score dataset, without having to repeatedly run the transaction prediction model, thereby improving data query efficiency. At the same time, the equity sensitivity score is obtained by calculating the transaction probability of users under different equity distribution conditions based on the transaction prediction model, which can more accurately characterize the actual sensitivity of users to equity. Based on the equity sensitivity score, the target population can be grouped, which can avoid or reduce the problem of rough characterization caused by simply relying on static labels, and achieve more refined population segmentation. On this basis, differentiated equity resource allocation can be carried out according to the sensitivity range of different sub-groups, which can not only improve the targeting and effectiveness of equity distribution, but also reduce resource waste while ensuring conversion rate, thereby improving the overall efficiency and use value of resource allocation.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this specification. Brief Description of the Drawings

[0020] FIG1 is a flowchart of an equity resource allocation method provided in an exemplary embodiment.

[0021] FIG2 is a schematic diagram of the output of a transaction prediction model provided in an exemplary embodiment.

[0022] FIG3 is a schematic diagram of an equity sensitivity curve provided in an exemplary embodiment.

[0023] FIG4 is a flowchart of determining whether an equity cap constraint is met provided in an exemplary embodiment.

[0024] FIG5 is a schematic diagram of the structure of an electronic device provided in an exemplary embodiment. Detailed Description of Embodiments

[0025] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of this specification. Rather, they are merely examples of apparatuses and methods consistent with some aspects of one or more embodiments of this specification as detailed in the appended claims.

[0026] It should be noted that in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the method may include more or fewer steps than described in this specification. Furthermore, a single step described in this specification may be broken down into multiple steps in other embodiments.The steps described in this specification may be combined into a single step in other embodiments.

[0027] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this specification are all information and data authorized by the user or fully authorized by all parties. The collection, use and processing of related data shall comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0028] Based on the problem of low utilization rate of rights and resources in related technologies, please refer to Figure 1. This specification provides a flowchart of a method for configuring rights and resources. This method can be executed by an electronic device, including but not limited to a server (such as a physical server, virtual server), a smartphone / mobile phone, a tablet computer, a personal digital assistant (PDA), a laptop computer, a desktop computer, a wearable device (e.g., a watch, glasses, gloves, headdress, pendant, etc.) or any other type of device. The method includes:

[0029] In S100, determining the target population to participate in the resource allocation activity.

[0030] This step is a preliminary step in the rights and benefits configuration. It can use preset user screening rules to identify specific groups of users who meet the goals of the resource allocation activity from the platform's total user base. The screening process can be based on basic user data, such as registration time, historical transaction records, account activity, etc. For example, when configuring repurchase rights for "old users who have made one transaction within 3 months but have not made any transactions in the past 30 days", the data screening logic will extract users who meet the conditions of "registered for more than 3 months + have made one transaction in the past 90 days + have not made any transactions in the past 30 days", forming a target population set, laying the foundation for accurate configuration of rights and benefits in the future.

[0031] In S102, the rights-sensitive scores of users in the target group are retrieved from the pre-stored rights-sensitive score dataset. The rights-sensitive scores of each user in the rights-sensitive score dataset are predicted using a preset transaction prediction model. The transaction prediction model includes two output branches: one output branch outputs the transaction probability of a user when no rights are issued, and the other output branch outputs the transaction probability of a user when rights are issued. The rights-sensitive score is used to characterize the difference between the transaction probability of a user when rights are issued and the transaction probability when no rights are issued.

[0032] In this step, only the user identifier of the target group, such as the user ID, needs to be used to match the corresponding rights-sensitive score from the pre-stored rights-sensitive score dataset. There is no need to repeatedly run the transaction prediction model, thus improving data query efficiency.

[0033] The pre-stored rights-sensitive score dataset is calculated and stored in advance by the preset transaction prediction model, and is not...Temporarily generated—The model is trained using historical user benefit distribution records and transaction behavior data (such as whether transactions occurred after benefit distribution and whether transactions occurred before benefit distribution). It possesses the ability to predict the degree of user response to benefit incentives. Referring to Figure 2, the input to this transaction prediction model consists of user input features, including but not limited to basic user characteristics (such as age, gender, occupation, etc.) and historical transaction features (such as historical bank card usage and spending history). This transaction prediction model has two output branches. One output branch outputs the user's transaction probability when no benefit is distributed, i.e., the natural transaction probability. The other output branch outputs the user's transaction probability when benefit is distributed, i.e., the benefit-incentivized transaction probability. The benefit sensitivity score is determined by the difference between the user's transaction probability when benefit is distributed and the transaction probability when benefit is not distributed, such as Benefit Sensitivity Score = Transaction Probability When Benefit is Distributed - Transaction Probability When Benefit is Not Distributed. This score characterizes the user's responsiveness to benefit incentives; the larger the difference, the higher the likelihood of the user making a transaction due to benefit incentives. This embodiment simultaneously considers both natural transaction probability and benefit-incentivized transaction probability, avoiding misjudgment based on a single dimension and providing a reliable data foundation for subsequent stratification.

[0034] The transaction prediction model is a machine learning model and requires training. This model can be trained on a training dataset, which includes two types of samples. One type includes the input features and corresponding transaction behavior labels (e.g., a first identifier indicating a completed transaction, and a second identifier indicating an incomplete transaction) of a first-type historical user who has received benefits. The other type includes the input features and corresponding transaction behavior labels (e.g., a first identifier indicating a completed transaction, and a second identifier indicating an incomplete transaction) of a second-type historical user who has not received benefits. Input features include basic user characteristics (e.g., age, gender, occupation, etc.) and transaction characteristics (e.g., bank card usage, historical spending, etc.) to enable the model to accurately predict the likelihood of user transactions.

[0035] It is understood that during the model training phase, the transaction probability output by the transaction prediction model typically requires post-processing steps such as threshold setting, probability discretization, or label alignment to convert continuous probability values ​​into discrete categories that can be directly compared with the supervision labels (transaction behavior labels), thereby completing the model's loss calculation and parameter update. In the actual inference stage, the electronic device can directly call the trained transaction prediction model and output the transaction probability of each user as input data for subsequent calculation of equity sensitivity score or other configuration indicators, without the need for post-processing steps such as label alignment.

[0036] By adopting a transaction prediction model with dual output branches, the transaction probability of each user can be obtained simultaneously under the same model structure.The transaction probabilities under both scenarios of issuing and not issuing benefits can more scientifically and precisely characterize the user's true sensitivity to benefit incentives, improve the accuracy and usability of sensitivity scores, and provide reliable data support for subsequent differentiated benefit allocation and budget control, thereby further optimizing the efficiency of benefit resource utilization and incentive effects.

[0037] Among them, the transaction prediction model can be modeled and trained separately for different resource allocation activities, different payment platforms, or different service scenarios. For example, if a user holds credit card 1 in institution 1 and credit card 2 in institution 2, and institution 1 initiates resource allocation activity 1 and institution 2 initiates resource allocation activity 2, then transaction prediction model 1 can be trained and generated based on the credit card service of institution 1 and resource allocation activity 1, and transaction prediction model 2 can be trained and generated based on the credit card service of institution 2 and resource allocation activity 2. Thus, for different payment platforms or institutions and different resource allocation activities, multiple corresponding benefit sensitivity score datasets can be generated and stored offline. In actual use, the electronic device can automatically select the matching transaction prediction model and the corresponding rights-sensitive score dataset according to the payment platform to which the target group belongs and the specific resource allocation activities to be participated in, so as to ensure that the generation and use of the sensitive score can more accurately reflect the real behavioral response characteristics of users in different scenarios.

[0038] In S104, according to the rights-sensitive scores of users in the target group, the target group is divided into at least two target subgroups, so that the rights-sensitive scores of users in different target subgroups belong to different rights-sensitive score ranges.

[0039] In this step, the electronic device can determine at least two rights-sensitive score ranges based on the rights-sensitive scores of users in the target group. Then, the electronic device can divide the target group into at least two target subgroups according to the rights-sensitive scores of users in the target group and at least two rights-sensitive score ranges. Grouping by sensitivity score range realizes "homogeneity within the same group and differentiation between different groups", providing a precise stratification basis for subsequent differentiated configuration, avoiding "one-size-fits-all" subsequent rights configuration, and allowing the rights needs of different subgroups to be accurately matched.

[0040] In one possible implementation, referring to Figure 3, the rights-sensitive scores of users in the target group can be sorted in descending order to obtain a rights-sensitive score curve; then, at least one target position point can be determined from the rights-sensitive score curve, the curvature of which is greater than the curvature of at least some other position points in the rights-sensitive score curve; finally, based on the rights-sensitive scores corresponding to at least one target position point, at least two rights-sensitive score ranges can be obtained. Specifically, the sensitivity scores can be sorted in descending order to form a curve (e.g., the horizontal axis is the sorting number, and the vertical axis is the rights-sensitive score), and then the second derivative of this curve can be calculated to find at least one curve with the largest second derivative (i.e., a significant change in curvature).The target location point serves as the dividing point between different ranges of sensitive rights scores. Using this as the boundary, the sensitive rights score curve is divided into multiple intervals, each interval corresponding to a range of sensitive rights scores.

[0041] In this implementation, by performing second derivative calculations on the sensitive rights score curve, the location point (inflection point) where the growth or decline trend changes most significantly in the curve is automatically detected. These inflection points often correspond to areas in the distribution of user sensitive scores where there are significant "hierarchical changes" or "group boundaries". Dividing based on inflection points can automatically adapt to the actual sensitive score distribution characteristics of the target population, rather than relying on manually set thresholds; accurately identifying the boundary between the most sensitive and least sensitive groups of rights helps to accurately allocate key resources; it is particularly applicable when there are abrupt changes, stepped distributions, or segmented characteristics in the sensitive scores, which can avoid the ambiguity of layering caused by "average division" and improve the accuracy and interpretability of layering. For example, when the user sensitive scores are densely clustered in the high-score segment, slowly declining in the middle segment, and rapidly collapsing in the low-score segment, the curvature inflection point can automatically locate these key change positions, achieve natural segmentation, and ensure that subsequent rights configuration is more targeted.

[0042] In another possible implementation, the rights and interests sensitivity scores of users in the target group can be sorted in descending order, and then the sorting results of the target group can be divided according to a fixed ratio. For example, the top 30% of users with the highest rights and interests sensitivity scores can be defined as the high-sensitivity range, the middle 50% as the medium-sensitivity range, and the remaining 20% ​​as the low-sensitivity range. The stratification ratio can be flexibly set according to specific operational goals and budget. This method is simple to implement and suitable for scenarios where the distribution of sensitivity scores is relatively smooth or where a balanced stratification interval is required. It can quickly match the preset population allocation strategy.

[0043] In yet another possible implementation, the electronic device can apply clustering analysis methods (such as K-means, hierarchical clustering, etc.) to the rights and interests sensitivity scores of the target group to automatically group users with similar sensitivity score characteristics into the same target subgroup, thereby obtaining several naturally formed sensitivity score ranges. This method does not rely on a preset threshold and can adaptively form a hierarchy based on the natural distribution characteristics of user sensitivity scores. It is suitable for situations where the distribution of sensitivity scores is complex, multi-peaked, or has local clusters, which helps to improve the scientific nature and operability of the hierarchy.

[0044] In S106, based on the range of rights sensitivity scores corresponding to at least two target subgroups, the target rights amount corresponding to at least two target subgroups is configured differently.

[0045] In this step, different subgroups obtain rights amounts that are appropriate to their own level of rights sensitivity scores, which avoids unnecessary cost expenditures for highly sensitive users due to excessive rights, and also avoids incentive failure for low-sensitive users due to insufficient rights, significantly improving the utilization rate of rights resources.

[0046] In one possible implementation, different amounts of benefits can be configured according to pre-configured rules based on the sensitivity range of the subgroups. For example, highly sensitive subgroups can be allocated more benefits to further amplify their transaction conversion potential by strengthening benefit incentives, ensuring that user groups with strong conversion promotion effects receive more sufficient benefit support; low-sensitive subgroups have weak response to benefits, and excessive benefits can easily lead to resource waste, so a lower target amount of benefits is allocated; medium-sensitive subgroups are allocated a middle-level amount of benefits. The configuration process automatically matches subgroups with benefit amounts through preset rules, without manual intervention.

[0047] In another possible implementation, considering that in various preset resource allocation tasks, benefit resources are a key incentive means, their configuration scheme is always strictly limited by the total amount of available resources, that is, there is a clear upper limit constraint on benefits. This constraint may originate from: the upper limit of the total amount of benefit resources predefined by the system or task setter; or, limited by the quantity of specific benefit resources that can be actually provided (such as coupons of specific face value, physical inventory); or, rigid control conditions set to ensure resource allocation efficiency and prevent resource consumption from exceeding the system's preset threshold. Once this constraint is exceeded, it will lead to uncontrolled resource consumption, failure of allocation schemes, and even affect the feasibility of overall resource planning.

[0048] Traditional rights and interests resource allocation methods often fail to achieve accurate and efficient allocation when faced with the above-mentioned rights and interests upper limit constraints. Some schemes adopt a simple average allocation strategy to strictly avoid the upper limit constraints. This "one-size-fits-all" approach leads to the waste of resources for groups with low rights and interests responsiveness, and fails to fully incentivize groups with high rights and interests responsiveness.

[0049] Therefore, in the implementation method that considers the rights and interests upper limit constraints, the electronic device can determine the benchmark rights and interests amount that meets the rights and interests upper limit constraints set by the resource allocation activity. For example, the benchmark rights and interests amount can be pre-calculated or set according to the budget of this resource allocation activity, the maximum amount of a single rights and interests distribution, or the rules of the platform, and used as a basic reference for subsequent allocation. "Rights and interests amount" refers to the specific value of rights and interests resources distributed to users, including but not limited to quantifiable parameters such as amount, discount rate, and number of points.

[0050] The electronic device can make differentiated adjustments to the benchmark rights and interests amount based on the rights and interests sensitive score ranges corresponding to at least two target subgroups to obtain the target rights and interests amount corresponding to at least two target subgroups. For example, a higher target benefit amount than the benchmark benefit amount can be allocated to a target subgroup with a higher benefit sensitivity score, while a target benefit amount equal to or lower than the benchmark benefit amount can be allocated to a target subgroup with a lower benefit sensitivity score, so that the adjusted target benefit amount matches the user's responsiveness to the benefits. By allocating different target benefit amounts according to the differences in benefit sensitivity scores among different target subgroups, see page 5 / 13 of the instruction manual, CN 121169473 A.This allows for the priority allocation of rights and benefits resources to user groups that have a greater impact on conversion, thereby improving the efficiency of rights and benefits resource utilization and overall conversion effect.

[0051] For example, the electronic device can determine the adjustment weight corresponding to each rights and benefits sensitivity score range, and the adjustment weight corresponding to each rights and benefits sensitivity score range is positively correlated with the rights and benefits sensitivity score in that range; then, based on the adjustment weight corresponding to each rights and benefits sensitivity score range, the benchmark rights and benefits amount is weighted and adjusted to obtain the target rights and benefits amount applicable to the target subgroup corresponding to that range. In this embodiment, by using an electronic device to determine the positively correlated adjustment weight based on each rights and benefits sensitivity score range and accordingly weighting and adjusting the benchmark rights and benefits amount, differentiated matching of incentive amounts for user groups with different sensitivity can be achieved. On the one hand, users with high sensitivity scores usually respond highly to incentive benefits. By allocating a higher amount of target benefits, their conversion probability can be further enhanced, and the overall effect of resource allocation activities can be improved. On the other hand, users with low sensitivity scores can be allocated a lower amount of target benefits, avoiding excessive benefits to people with low conversion probability and reducing resource waste. This weighted adjustment mechanism is based on the objective distribution of sensitivity scores and the preset weight mapping rules, which improves the accuracy and scientific nature of incentive configuration and ensures a higher input-output ratio under the same benefit budget.

[0052] The adjustment weight corresponding to each range of sensitivity scores can be obtained by mapping the statistical value of the sensitivity scores of the target subgroup corresponding to the range of sensitivity scores through a pre-set mapping relationship between sensitivity scores and adjustment weights.

[0053] For example, the mapping relationship between sensitivity scores and adjustment weights can be a pre-defined mapping table, such as linking sensitivity scores to fixed weights by intervals, which is simple, efficient, and easy to configure and maintain.

[0054] For example, the mapping relationship between the rights and interests sensitivity score and the adjustment weight can be expressed as a formulaic function (such as a linear function, a piecewise function, or a logarithmic function). This function automatically calculates and adjusts the weight in real time based on the input sensitivity score statistics, supporting more granular dynamic adjustment.

[0055] For example, an electronic device can first stratify the target population into three rights and interests sensitivity score ranges: the first layer (high sensitivity score range): the user's rights and interests sensitivity score is between 0.8 and 1.0; the second layer (medium sensitivity score range): the user's rights and interests sensitivity score is between 0.5 and 0.8; and the third layer (low sensitivity score range): the user's rights and interests sensitivity score is between 0 and 0.5.

[0056] At this time, the electron can, according to a preset mapping relationship, such as the higher the sensitivity score, the greater the adjustment weight, substitute the representative statistical value (e.g., mean or median) corresponding to the sensitivity score range into the mapping relationship. For example: the high sensitivity stratum corresponds to a mean of 0.9, and the adjusted weight after mapping is 1.2; the medium sensitivity stratum corresponds to a mean of 0.65, and the adjusted weight after mapping is 1.0; the low sensitivity stratum...The corresponding mean is 0.3, and the adjusted weight after mapping is 0.8. Then, based on these adjusted weights, the electronic device performs a weighted adjustment on the baseline benefit amount (e.g., a 10 yuan red envelope): the target benefit amount corresponding to the high-sensitivity subgroup = 10 × 1.2 = 12; the target benefit amount corresponding to the medium-sensitivity subgroup = 10 × 1.0 = 10; and the target benefit amount corresponding to the low-sensitivity subgroup = 10 × 0.8 = 8; thereby matching a more reasonable benefit investment level for different groups.

[0057] For example, after obtaining the target benefit amounts corresponding to at least two target subgroups, the electronic device will determine whether the target benefit amounts corresponding to at least two target subgroups meet the set benefit upper limit constraint. If the target benefit amounts corresponding to at least two target subgroups do not meet the benefit upper limit constraint, the electronic device can correct the target benefit amounts of at least one target subgroup until the target benefit amounts corresponding to at least two target subgroups meet the benefit upper limit constraint. For example, during the correction process, the benefit amount of the target subgroup with lower benefit sensitivity can be reduced first to ensure the incentive allocation for the high-sensitivity group. After correction, the electronic device will re-verify whether the constraint is met. If it is still not met, the correction will be repeated until the target amount of rights meets the upper limit constraint.

[0058] By automatically correcting the target amount of rights under the upper limit constraint, the distribution of rights is ensured to be within a controllable range, avoiding excessive consumption of resources, while still maximizing the coverage of user groups with high responsiveness to resource allocation activities. This helps to improve the conversion rate and the rationality of budget execution. A closed-loop process of "benchmark determination - difference adjustment - verification and correction" is constructed with the upper limit constraint of rights as the core, which fundamentally reduces the redundancy of invalid rights distribution and improves the utilization efficiency of limited rights resources at the technical level.

[0059] For example, please refer to Figure 4. The electronic device can obtain the statistical weight of the target rights amount corresponding to each target subgroup (S400). The statistical weight is used to characterize the relative influence of each target subgroup in the overall rights allocation. Then, based on the target rights amount corresponding to at least two target subgroups and the statistical weight, a weighted statistical value is obtained (S402). Then, it is determined whether the weighted statistical value is greater than the rights upper limit constraint (S404). If the weighted statistical value is greater than the rights upper limit constraint, the electronic device corrects the target rights amount of at least one target subgroup (S406) and performs weighted statistics again until the weighted statistical value obtained again is not greater than the rights upper limit constraint.

[0060] In this embodiment, the relationship between the hierarchical differentiated rights configuration and the rights upper limit constraint can be dynamically balanced to ensure that even if different target subgroups are allocated different target rights amounts, the rights upper limit constraint requirement will not be exceeded, which meets the requirements.Cost controllability requirement; if the weighted statistical benefit amount exceeds the upper limit, the electronic device can automatically correct the target benefit amount of some sub-groups and repeatedly execute the weighted statistics until the constraint is met, avoiding multiple manual attempts and repeated adjustments, and improving the automation and real-time performance of strategy adjustment; this process is based on data-driven weighted statistics and iterative verification, ensuring the scientific nature, executability and stability of benefit configuration, and helping to quickly implement it in actual large-scale groups.

[0061] For example, suppose an e-commerce platform plans to issue full-reduction red envelopes to target users, and the operation team sets the unit benefit upper limit constraint to 12 yuan / single transaction, and pre-determines the benchmark benefit amount to 12 yuan / single transaction.

[0062] Then, the electronic device divides the target group into: high-sensitivity group A (high conversion potential), medium-sensitivity group B (moderate conversion potential), and low-sensitivity group C (low conversion potential) according to the above division method. According to the sensitivity range, the system assigns adjustment weights to different groups: A: 1.1; B: 1.0; C: 0.8. Therefore: A's initial target equity amount = 12 × 1.1 = 13.2 yuan; B's initial target equity amount = 12 × 1.0 = 12 yuan; C's initial target equity amount = 12 × 0.8 = 9.6 yuan.

[0063] Assume that the statistical weights of the target equity amounts corresponding to each target subgroup are: A's statistical weight = 0.5; B's statistical weight = 0.3; C's statistical weight = 0.2. Therefore, the weighted statistical equity amount = 13.2 × 0.5 + 12 × 0.3 + 9.6 × 0.2 = 6.6 + 3.6 + 1.92 = 12.12 yuan.

[0064] At this time, it is found that the weighted statistical equity amount exceeds the equity upper limit constraint of 12 yuan. The electronic device will perform correction. For example, it can prioritize lowering the equity amount for the low-sensitivity subgroup C from 9.6 yuan to 8 yuan, and then recalculate the weighted statistical equity amount to 11.8 yuan, which satisfies the equity upper limit constraint. By using weighted statistics and automatic correction, not only is the upper limit constraint of rights and interests met, but also priority is given to high-response groups, avoiding resource waste and helping to improve the efficiency of deployment.

[0065] In one possible implementation, the statistical weight of the target rights and interests corresponding to the target subgroup can be determined by the estimated transaction probability of the target subgroup. For example, the estimated transaction probability of the target subgroup is determined by the following method: obtaining the transaction information of the historical groups participating in the resource allocation activity, including the rights and interests sensitive sub-intervals to which the users in the historical groups belong and the transaction probability corresponding to each rights and interests sensitive sub-interval; if the target subgroup corresponds to a single rights and interests sensitive sub-interval, the transaction probability corresponding to that rights and interests sensitive sub-interval is determined as the estimated transaction probability of the target subgroup; if the target subgroup involves at least two rights and interests sensitive sub-intervals, the estimated transaction probability of the target subgroup is calculated based on the transaction probabilities corresponding to the at least two rights and interests sensitive sub-intervals respectively, and combined with the proportion of the target subgroup in each sensitive sub-interval.

[0066] In this embodiment, by setting the statistical weight of the target subgroup to the estimated transaction probability, the proportion of the overall equity budget for different sensitivity stratification groups can more reasonably reflect the true conversion potential. This statistical weight directly reflects the actual possible response level of the target subgroup to the resource allocation activity, rather than simply allocating resources according to the average size of the group, thus avoiding excessive waste of budget for low-response groups. This dynamic weight setting based on historical response performance reduces the arbitrary distribution strategy of equity, and helps to achieve resource allocation according to potential and equity adjustment according to priority, thereby achieving a higher overall conversion rate under budget constraints.

[0067] In another possible implementation, the statistical weight of the target equity amount corresponding to the target subgroup is determined by the estimated number of transactions of the target subgroup. The estimated number of transactions of the target subgroup includes the product between the estimated transaction probability of the target subgroup and the total number of the target subgroup. By determining the statistical weight based on the estimated number of transactions of the target subgroup, the equity allocation can be based not only on the proportion of the number of subgroups, but also on the actual transaction potential of each subgroup for weighting. By comprehensively considering the size and potential transaction intentions of subgroups, the overall weighted statistical results can more realistically reflect the possible contribution of different subgroups to the overall transaction conversion. Therefore, the final equity allocation scheme, under the premise of meeting the overall equity upper limit constraint, can realize the allocation of more equity to high-potential groups and the reasonable control of equity resources for low-potential groups, thereby improving the efficiency of equity allocation and the overall conversion effect.

[0068] For example, subgroup A has 10,000 people and an estimated conversion rate of 0.1 → product = 1000; subgroup B has 5,000 people and an estimated conversion rate of 0.3 → product = 1500; after normalizing the two products, the normalized statistical weight of subgroup A is 0.4 and the normalized statistical weight of subgroup B is 0.6. The statistical weights obtained after normalization can be directly used to calculate the weighted average of the target equity amount, ensuring the reasonable reflection of both the target subgroup size and the estimated conversion rate, and making the equity allocation more accurate. The larger the statistical weight, the more numerous the target subgroup is, and the more likely it is to convert; therefore, its "contribution weight" in the overall equity allocation is greater. In the weighted calculation stage, the statistical weight directly affects the result of the weighted average equity amount, ensuring that more valuable and high-potential subgroups receive a matching equity incentive share.

[0069] In another possible implementation, the statistical weight of the target equity amount corresponding to the target subgroup can be determined based on the population size of the target subgroup, that is, the proportion of the target subgroup in the entire target population. The more people in the subgroup, the higher the statistical weight.

[0070] In some embodiments, the electronic device determines that the weighted statistical equity amount does not meet the preset equity upper limit constraint.Then, based on the rights sensitivity score range corresponding to at least two target subgroups and the preset rights amount lower limit of each target subgroup, at least one target subgroup to be adjusted can be determined from at least two target subgroups.

[0071] For example, the electronic device can prioritize excluding target subgroups whose target rights amount is equal to or less than their corresponding preset rights amount lower limit from at least two target subgroups to avoid lowering the incentive amount below the guarantee level. After that, the electronic device can determine the target subgroups with lower rights sensitivity scores as the objects to be adjusted from the remaining target subgroups. That is, the rights sensitivity scores of users in the target subgroups to be adjusted are lower than the rights sensitivity scores of users in other target subgroups in the remaining target subgroups. A lower rights sensitivity score indicates that the target subgroup has a relatively low response to rights incentives or conversion potential. This mechanism ensures that the reduction of incentives for low-sensitivity users will not affect users who are already at the rights guarantee lower limit by first excluding the critical lower limit group.

[0072] Finally, the electronic device can gradually reduce the target rights amount of the target subgroup without lowering it below the preset rights amount lower limit of the target subgroup to be adjusted, until the entire weighted statistical rights amount meets the rights upper limit constraint. This achieves priority reduction of incentive resources for target subgroups with lower equity sensitivity scores, thus optimizing the allocation efficiency of incentive resources and improving the overall conversion effect. This dynamic correction scheme makes the whole process more automated and adjustable, and can accurately match the budget in multiple iterations, significantly reducing the risk of over-budget or inefficient deployment.

[0073] In other implementations, after the electronic device determines that the weighted statistical equity amount does not meet the preset equity upper limit constraint, it can determine the normalization coefficient based on the ratio between the weighted statistical equity amount and the equity upper limit constraint, and then normalize the target equity amount corresponding to each target subgroup based on the normalization coefficient. This scheme achieves global synchronous scaling through the normalization coefficient, which is simple and efficient, suitable for large-scale multi-subgroup scenarios, and does not require individual judgment and adjustment. Normalization can ensure that the relative incentive intensity ratio between each group remains consistent, reducing the cost of manual parameter adjustment.

[0074] For example, assuming the weighted statistical benefit amount = 15 yuan / transaction; the preset benefit upper limit constraint = 12 yuan / transaction; then the normalization coefficient = 12 / 15 = 0.8. If the original target benefit amount for subgroup A = 14 yuan; the target benefit amount for subgroup B = 16 yuan, then after normalization: the corrected benefit amount for A = 14 × 0.8 = 11.2 yuan; the corrected benefit amount for B = 16 × 0.8 = 12.8 yuan.

[0075] In some embodiments, after the electronic device determines that the target benefit amounts corresponding to at least two target subgroups meet the preset benefit upper limit constraint, the operator can adjust the benefit amount for each target subgroup based on the target benefit amount corresponding to each target subgroup.Different target subgroups execute differentiated resource allocation activities to match them, thereby improving the reach and conversion effect of different groups. During the actual implementation of resource allocation activities, electronic devices can obtain the actual transaction probability or actual number of transactions of at least two target subgroups in real time or periodically, so as to reflect the real response effect of different groups to the current rights and interests configuration.

[0076] Further, when the electronic device finds that the actual transaction probability (or actual number of transactions) of one or more target subgroups is significantly higher than expected, it can adjust the corresponding target rights and interests amount based on the actual performance of the subgroup to avoid excessive resource investment. Conversely, if it finds that the actual transaction probability (or actual number of transactions) of a target subgroup is significantly lower than expected, it can increase the target rights and interests amount of the target subgroup to enhance the incentive for the group and promote subsequent transaction conversion.

[0077] In one possible implementation, the electronic device can determine a dynamic correction coefficient based on the ratio between the actual transaction probability and the estimated transaction probability. This correction coefficient can be used to proportionally amplify or reduce the original target rights and interests amount, thereby generating an updated target rights and interests amount.

[0078] After the update is completed, the electronic device can perform weighted statistics again based on the adjusted target equity amount and determine whether the updated target equity amount still meets the equity upper limit constraint. If not, the electronic device can further modify the target equity amount for some target subgroups until the adjusted target equity amount meets the preset equity upper limit constraint, thereby forming a closed-loop optimization.

[0079] By dynamically modifying the target equity amount based on the actual transaction effect during the execution of resource allocation activities, the waste of resources or insufficient conversion caused by a single static configuration can be effectively avoided, the matching degree and flexibility of equity allocation for different sensitive groups can be improved, and the balance between accurate equity placement and controllable costs can be achieved. At the same time, combined with the dynamic verification of the equity upper limit constraint, the total budget can be further guaranteed not to be exceeded, enhancing the sustainability of the overall strategy and the ability to maximize returns.

[0080] In some optional embodiments, in order to verify the effect of the rights-sensitive tiered configuration strategy, the operator can also set up an additional benchmark group without differentiated rights adjustments in the same resource allocation activity. This benchmark group can correspond to a pre-set benchmark rights amount, and the same or similar resource allocation activities are performed on this benchmark group for effect comparison analysis with other subgroups. By comparing the conversion performance, rights utilization efficiency and other indicators of different groups, data support can be provided for subsequent strategy optimization, further verifying the superiority of tiered configuration and the rationality of resource allocation.

[0081] The various technical features in the above embodiments can be arbitrarily combined, as long as there is no conflict or contradiction between the combinations of features. However, due to space limitations, they are not described one by one. Therefore, the arbitrary combination of various technical features in the above embodiments is also within the scope of this specification.

[0082] In some embodiments, this specification also provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor implements the method described in specification 9 / 13 pages 12 CN 121169473 A by running the executable instructions.

[0083] FIG5 is a schematic structural diagram of a device provided in an exemplary embodiment. As shown in FIG5, the device 500 mainly consists of a communication interface 502, a user interface 504, a processor 506, and a data storage 508. These components are interconnected and communicate with each other through a system bus, network, or other connection mechanism 510. The communication interface 502 enables the device 500 to communicate with other devices, access networks, and transmission networks through analog or digital modulation. For example, the communication interface 502 may include a chipset and an antenna for wireless communication with a radio access network or access point. In addition, the communication interface 502 may also be a wired interface such as Ethernet, token ring, or USB port, or a wireless interface such as Wifi, Bluetooth, Global Positioning System (GPS), or wide area wireless interface (e.g., WiMAX or LTE). Of course, the communication interface 502 may also support other forms of physical layer interfaces and standard or proprietary communication protocols. The communication interface 502 may also include multiple physical communication interfaces, such as Wi-Fi interfaces, Bluetooth interfaces, and wide area wireless interfaces.

[0084] The user interface 504 includes receiving user input and providing output to the user. Therefore, the user interface 504 may include input components such as keypads, keyboards, touch-sensitive or presence-sensitive panels, computer mice, trackballs, joysticks, microphones, still cameras, and video cameras, and may also include output components such as display screens (which may be combined with touch-sensitive panels), CRTs, LCDs, LEDs, displays using DLP technology, printers, and other similar devices known or developed in the future. The user interface 504 may also generate auditory output through speakers, speaker jacks, audio output ports, audio output devices, headphones, and other similar devices known or developed in the future. In some embodiments, the user interface 504 may include software, circuitry, or other forms of logic capable of transmitting data to and receiving data from external user input / output devices. Alternatively, device 500 may support remote access from other devices via communication interface 502 or another physical interface (not shown). User interface 504 may be configured to receive user input, the position and movement of which may be indicated by an indicator or cursor described herein. User interface 504 may also be configured as a display device for rendering or displaying text fragments.

[0085] Processor 506 may comprise one or more general-purpose processors and / or dedicated processors.

[0086] Data storage 508 may include one or more volatile and / or non-volatile storage components and may be integrated wholly or partially with processor 506. Data storage 508 may include removable and non-removable components.

[0087] Processor 506 is capable of executing program instructions 518 (e.g., compiled or uncompiled program logic and / or machine code) stored in data storage 508 to perform various functions herein. Data storage 508 may include a non-transitory computer-readable medium having program instructions stored thereon that, when executed by device 500, enable device 500 to perform any methods, processes, or functions disclosed in this specification and / or drawings. Execution of program instructions 518 by processor 506 may result in processor 506 using data 512.

[0088] For example, program instructions 518 may include an operating system 522 (e.g., an operating system kernel, device drivers, and / or other modules) installed on device 500 and one or more application programs 520 (e.g., a browser, social application, or game application). Similarly, data 512 may include operating system data 516 and application data 514. Operating system data 516 is primarily accessible to operating system 522, while application data 514 is primarily accessible to one or more applications 520. Application data 514 may reside in a file system visible or hidden from the user of device 500.

[0089] Application 520 may communicate with operating system 522 through one or more application programming interfaces (APIs). These APIs facilitate application 520 reading and / or writing application data 514, transmitting or receiving information via communication interface 502, receiving or displaying information on user interface 504, etc.

[0090] In some terms, application 520 may be simply referred to as "app". Furthermore, application 520 may be downloaded to device 500 through one or more online app stores or app markets. However, applications may also be installed on device 500 by other means, such as through a web browser or a physical interface on device 500 (e.g., a USB port). Instruction Manual 10 / 13 pages 13 CN 121169473 A

[0091] In some embodiments, the rights allocation device can be applied to the device shown in FIG5 to implement the technical solution of this specification. The rights resource allocation device may include:

[0092] a target population determination module, used to determine the target population to participate in the resource allocation activity.

[0093] a rights sensitivity score query module, used to query the rights sensitivity scores of users in the target population from a pre-stored rights sensitivity score dataset. The rights sensitivity scores of each user in the rights sensitivity score dataset are predicted using a preset transaction prediction model; the transaction prediction model includes two output branches, one of which is used to output the user's...The transaction probability under the condition of no rights being issued, and another output branch is used to output the transaction probability of the user under the condition of rights being issued. The rights sensitivity score is used to characterize the difference between the transaction probability of the user under the condition of rights being issued and the transaction probability under the condition of no rights being issued.

[0094] The target audience segmentation module is used to divide the target audience into at least two target sub-groups according to the rights sensitivity scores of users in the target audience, so that the rights sensitivity scores of users in different target sub-groups belong to different rights sensitivity score ranges.

[0095] The target rights amount configuration module is used to differentiate the target rights amount corresponding to at least two target sub-groups based on the rights sensitivity score ranges corresponding to at least two target sub-groups.

[0096] In one implementation, the target population segmentation module is specifically used to sort the rights-sensitive scores of users in the target population in descending order to obtain a rights-sensitive score curve; determine at least one target position point from the rights-sensitive score curve, wherein the curvature of the target position point is greater than the curvature of at least some other position points in the rights-sensitive score curve; obtain at least two rights-sensitive score ranges based on the rights-sensitive scores corresponding to the at least one target position point; and divide the target population into at least two target sub-populations based on the rights-sensitive scores of users in the target population and the at least two rights-sensitive score ranges.

[0097] In one implementation, the target rights quantity configuration module is specifically used to determine the benchmark rights quantity that satisfies the rights upper limit constraint set by the resource allocation activity; make differential adjustments to the benchmark rights quantity based on the rights-sensitive score ranges corresponding to the at least two target sub-populations to obtain the target rights quantity corresponding to the at least two target sub-populations; if the target rights quantity corresponding to the at least two target sub-populations does not satisfy the rights upper limit constraint, correct the target rights quantity of at least one target sub-population until the target rights quantity corresponding to the at least two target sub-populations satisfies the rights upper limit constraint.

[0098] In one implementation, the target equity allocation module is specifically used to obtain the statistical weight of the target equity allocation corresponding to each target subgroup; perform weighted statistics based on the target equity allocation and statistical weight corresponding to at least two target subgroups respectively to obtain a weighted statistical equity allocation; if the weighted statistical equity allocation is greater than the equity upper limit constraint, the target equity allocation of at least one target subgroup is corrected, and weighted statistics are performed again until the weighted statistical equity allocation obtained again is not greater than the equity upper limit constraint.

[0099] In one implementation, the statistical weight of the target equity allocation corresponding to the target subgroup is determined by the estimated transaction probability of the target subgroup; or, the statistical weight of the target equity allocation corresponding to the target subgroup is determined by the estimated number of transactions of the target subgroup, wherein the estimated number of transactions of the target subgroup includes the product between the estimated transaction probability of the target subgroup and the total number of target subgroups.

[0100] In one implementation, a predicted transaction probability acquisition module is also included, which is used to acquire transaction information of historical groups participating in resource allocation activities. The transaction information includes the rights-sensitive sub-segments to which users in the historical groups belong and the transaction probabilities corresponding to each rights-sensitive sub-segment. If the target sub-group corresponds to a single rights-sensitive sub-segment, the transaction probability corresponding to that rights-sensitive sub-segment is determined as the predicted transaction probability of the target sub-group. If the target sub-group involves at least two rights-sensitive sub-segments, the predicted transaction probability of the target sub-group is calculated based on the transaction probabilities corresponding to the at least two rights-sensitive sub-segments respectively, and in combination with the proportion of the target sub-group in each sensitive sub-segment as shown on pages 11 / 13 of the specification, CN 121169473 A.

[0101] In one implementation, the target rights amount configuration module is specifically used to determine the adjustment weight corresponding to each rights-sensitive sub-segment. The adjustment weight corresponding to each rights-sensitive sub-segment is positively correlated with the rights-sensitive sub-segment in that rights-sensitive sub-segment. Based on the adjustment weight corresponding to each rights-sensitive sub-segment, the benchmark rights amount is weighted and adjusted to obtain the target rights amount applicable to the target sub-group corresponding to that range.

[0102] In one implementation, the adjustment weight corresponding to each rights-sensitive score range is obtained by mapping the statistical value of the rights-sensitive score of the target subgroup corresponding to the rights-sensitive score range through a pre-set mapping relationship between rights-sensitive scores and adjustment weights.

[0103] In one implementation, the target rights amount configuration module is specifically used to determine at least one target subgroup to be adjusted from at least two target subgroups based on the rights-sensitive score ranges corresponding to at least two target subgroups and the preset rights amount lower limit of each target subgroup; and to lower the target rights amount of the target subgroup to be adjusted if it is not lower than the preset rights amount lower limit of the target subgroup to be adjusted.

[0104] In one implementation, the target rights amount configuration module is specifically used to exclude target subgroups whose target rights amount is equal to or less than their corresponding preset rights amount lower limit from at least two target subgroups; and to determine at least one target subgroup to be adjusted from the remaining target subgroups, wherein the rights-sensitive score of the users in the target subgroup to be adjusted is lower than the rights-sensitive score of the users in other target subgroups in the remaining target subgroups.

[0105] In one implementation, the target equity allocation module is specifically used to determine a normalization coefficient based on the ratio between the weighted statistical equity allocation and the equity ceiling constraint; and to normalize the target equity allocation corresponding to each target subgroup based on the normalization coefficient.

[0106] The implementation process of the functions and roles of each module in the above device is detailed in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0107] Based on the same concept as the above method, this specification also provides a computer-readable storage medium thereon.The storage contains computer instructions that, when executed by a processor, implement the steps of the method as described in any of the above embodiments.

[0108] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology for information storage. The information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transferable medium that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0109] Based on the same concept as the above method, this specification also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the method as described in any of the above embodiments.

[0110] It will be understood by those skilled in the art that:

[0111] In this specification, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Specification 12 / 13 pages 15 CN 121169473 A Without further limitations, it is not excluded that other identical or equivalent elements may be present in a process, method, product, or apparatus that includes said elements.

[0112] In this specification, "a," "an," and "the" do not specifically refer to the singular and may also include the plural.

[0113] In this specification, ordinal numbers such as first and second do not necessarily indicate order; they are often used to distinguish objects. For example, "first server" and "second server" usually refer to two servers. To distinguish between these two servers, they are described as "first server" and "second server." Of course, sometimes these two servers may be the same server.

[0114] In this specification, unless explicitly stated otherwise, "receiving and sending data" does not necessarily mean direct receiving and sending, but can be indirect receiving and sending. For example, A receiving data sent by B can be understood as A directly receiving data sent by B.The data sent can also be understood as A indirectly receiving data sent by B through other entities such as C. Similarly, B sending data to A can be understood as B sending data directly to A, or it can be understood as B indirectly sending data to A through other entities such as C. Here, C can be one entity, or it can be two or more entities.

[0115] In this specification, unless explicitly stated otherwise, the relationship between structures can be a direct relationship or an indirect relationship. For example, when describing "A is connected to B", unless it is explicitly stated that A is directly connected to B, it should be understood that A can be directly connected to B or indirectly connected to B; as another example, when describing "A is above B", unless it is explicitly stated that A is directly above B (AB is adjacent and A is above B), it should be understood that A can be directly above B, or A can be indirectly above B (AB is separated by other elements, and A is above B). And so on.

[0116] This specification uses specific terms to describe the embodiments of this specification. The terms "an embodiment," "an embodiment," and / or "some embodiments" refer to a feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that "an embodiment," "an embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this specification do not necessarily refer to the same embodiment. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples, without contradiction.

[0117] Although one or more embodiments of this specification provide method steps as described in the embodiments or flowcharts, it is understood that the order of steps listed in the embodiments or flowcharts is merely one possible execution order among many steps and does not represent a unique execution order. Therefore, when the claims involve method steps, adjustments to the order of such steps, or parallel execution between steps, are also within the scope of the claims. Specification page 13 / 13, 16 CN 121169473 A, Figure 1, Figure 2; Specification drawing page 1 / 3, 17 CN 121169473 A, Figure 3, Figure 4; Specification drawing page 2 / 3, 18 CN 121169473 A, Figure 5; Specification drawing page 3 / 3, 19 CN 121169473 A, Abstract: One or more embodiments of the specification provide a benefit resource allocation method and device, a storage medium, and a program product.The benefit allocation method includes: determining a target crowd to participate in a resource allocation activity; acquiring a benefit sensitivity score for each user in the target crowd from a pre-stored benefit sensitivity score dataset, where the benefit sensitivity score is calculated by a preset transaction prediction model, and the transaction prediction model includes two output branches: one branch is used for outputting a transaction probability of the user under the condition that the benefit is not issued, and the other branch is used for outputting a transaction probability of the user under the condition that the benefit is issued, where the benefit sensitivity score is used for representing the difference between the two transaction probabilities; dividing the target crowd into at least two target sub-crowds according to the benefit sensitivity score of the user, and then differentially allocating target benefit amount for the at least two target sub-crowds, so thatmore accurate resource release and distribution are realized.

Claims

1. A method for allocating equity resources, comprising: Identify the target population for the resource allocation activities; From the pre-stored data set of rights-sensitive scores, retrieve the rights-sensitive scores of users in the target population; The rights-sensitive score of each user in the rights-sensitive score dataset is predicted using a preset transaction prediction model. The transaction prediction model includes two output branches, one of which outputs the transaction probability of a user when no rights are issued, and the other output branch outputs the transaction probability of a user when rights are issued. The rights-sensitive score is used to characterize the difference between the transaction probability of a user when rights are issued and the transaction probability when no rights are issued. Based on the rights and interests sensitivity scores of users in the target group, the target group is divided into at least two target subgroups, such that the rights and interests sensitivity scores of users in different target subgroups belong to different rights and interests sensitivity score ranges. Based on the rights-sensitive sub-ranges corresponding to the at least two target sub-groups, the target rights amount corresponding to the at least two target sub-groups is configured differently.

2. The method according to claim 1, wherein dividing the target population into at least two target sub-groups based on the rights sensitivity scores of users within the target population includes: The rights and interests sensitivity scores of users in the target group are sorted in descending order to obtain the rights and interests sensitivity score curve; At least one target location point is determined from the equity-sensitive sub-curve, wherein the curvature of the target location point is greater than the curvature of at least some other location points in the equity-sensitive sub-curve; Based on the equity sensitivity scores corresponding to the at least one target location point, at least two equity sensitivity score ranges are obtained; Based on the rights and interests sensitivity scores of users in the target population and the ranges of the at least two rights and interests sensitivity scores, the target population is divided into at least two target sub-groups.

3. The method according to claim 1, wherein the step of differentially configuring the target equity amount corresponding to the at least two target subgroups based on the equity sensitivity ranges corresponding to the at least two target subgroups includes: Determine the baseline equity amount that satisfies the equity cap constraints set for resource allocation activities; Based on the rights-sensitive sub-ranges corresponding to the at least two target sub-groups, the benchmark rights amount is adjusted differentially to obtain the target rights amount corresponding to the at least two target sub-groups. If the target equity amounts corresponding to the at least two target subgroups do not meet the equity upper limit constraint, the target equity amounts of at least one target subgroup shall be adjusted until the target equity amounts corresponding to the at least two target subgroups meet the equity upper limit constraint.

4. The method according to claim 3, wherein if the target equity amounts corresponding to the at least two target subgroups do not meet the equity upper limit constraint, the target equity amounts of at least one target subgroup are adjusted until the target equity amounts corresponding to the at least two target subgroups meet the equity upper limit constraint, comprising: Obtain the statistical weights of the target equity amounts corresponding to each target subgroup; The weighted statistical equity amount is obtained by performing weighted statistics based on the target equity amount corresponding to the at least two target subgroups and the statistical weights. If the weighted statistical equity amount is greater than the equity upper limit constraint, the target equity amount of at least one of the target subgroups is corrected, and weighted statistics are performed again until the weighted statistical equity amount obtained again is not greater than the equity upper limit constraint.

5. The method according to claim 4, wherein the statistical weight of the target equity corresponding to the target subgroup is determined by the estimated transaction probability of the target subgroup; Alternatively, the statistical weight of the target equity corresponding to the target subgroup is determined by the estimated number of transactions of the target subgroup, wherein the estimated number of transactions of the target subgroup includes the product between the estimated transaction probability of the target subgroup and the total number of the target subgroup. The estimated transaction probability of the target subgroup is determined in the following way: Obtain transaction information of historical participants in the resource allocation activity, including the rights-sensitive sub-segments to which users in the historical group belong and the transaction probability corresponding to each rights-sensitive sub-segment; If the target subgroup corresponds to a single equity-sensitive segment, the transaction probability corresponding to that equity-sensitive segment is determined as the estimated transaction probability of the target subgroup. If the target subgroup involves at least two equity-sensitive intervals, the estimated transaction probability of the target subgroup is calculated based on the transaction probabilities corresponding to the at least two equity-sensitive intervals and the proportion of the target subgroup in each sensitive interval.

6. The method according to claim 3, wherein the step of differentially adjusting the benchmark equity amount based on the equity sensitivity range corresponding to the at least two target subgroups to obtain the target equity amount corresponding to the at least two target subgroups includes: Determine the adjustment weight corresponding to each equity sensitivity score range. The adjustment weight corresponding to each equity sensitivity score range is positively correlated with the equity sensitivity score in that equity sensitivity score range. Based on the adjustment weights corresponding to each sensitive rights and interests range, the benchmark rights and interests amount is adjusted by weighting to obtain the target rights and interests amount applicable to the target subgroups corresponding to that range.

7. The method according to claim 4, wherein modifying the target equity amount for at least one of the target subgroups comprises: Based on the rights-sensitive sub-ranges corresponding to the at least two target sub-groups and the preset rights amount lower limit of each target sub-group, at least one target sub-group to be adjusted is determined from the at least two target sub-groups. If the target rights amount of the target sub-group to be adjusted is not lower than the preset rights amount lower limit of the target sub-group to be adjusted, the target rights amount of the target sub-group to be adjusted is reduced. Alternatively, a normalization coefficient can be determined based on the ratio between the weighted statistical equity amount and the equity upper limit constraint, and the target equity amount corresponding to each target subgroup can be normalized based on the normalization coefficient.

8. An electronic device, comprising: processor; A memory for storing processor-executable instructions; wherein the processor implements the steps of the method as described in any one of claims 1 to 7 by executing the executable instructions.

9. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the steps of the method as claimed in any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 7.

Images