Right and interest distribution method and device and electronic equipment

Abstract

The invention provides a right and interest distribution method and device and electronic equipment, and the method comprises the steps: obtaining the historical operation behavior data of each user for each right and interest in a current right and interest database, inputting each historical operation behavior data into a trained user feature prediction model, obtaining the user feature data of each user, and obtaining the user feature data of each user; the user feature data comprises preference probabilities of the corresponding users for the rights and interests, determining an objective function based on the preference probabilities and the distribution costs of the rights and interests, determining to-be-distributed rights and interests of the users from a rights and interests database according to the objective function and preset constraint conditions, and distributing the to-be-distributed rights and interests to the corresponding users. Automatic allocation of the user rights and interests is achieved, the rights and interests are allocated to all the users based on the preference probability of the users for all the rights and interests, and the suitability of rights and interests allocation is improved.

Description

Technical Field

[0001] The present application relates to the field of machine learning technology, and in particular to a rights distribution method, device and electronic device. Background Art

[0002] In the digital economy era, many apps need to recommend benefits to users. The distribution of digital benefits such as e-books, music, videos and other content has become the focus of attention of enterprises and users. Therefore, how to automatically recommend appropriate benefits to users has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a rights distribution method, device and electronic device to solve the above technical problems.

[0004] In one aspect, a method for distributing equity is provided, the method comprising:

[0005] Obtain the historical operation behavior data of each user on each equity in the current equity database;

[0006] Inputting each of the historical operation behavior data into the trained user feature prediction model to obtain user feature data of each of the users; the user feature data includes the preference probability of the corresponding user for each of the rights and interests;

[0007] Determining an objective function based on each of the preference probabilities and the distribution cost of each of the benefits;

[0008] Determining the rights and interests to be distributed to each of the users from the rights and interests database according to the objective function and preset constraints;

[0009] Distribute the rights to be distributed to the corresponding users.

[0010] In one embodiment, obtaining the historical operation behavior data of each user on each right in the current right database includes:

[0011] Collect the original historical operation behavior data of each user on each right in the current right database;

[0012] The original historical operation behavior data are preprocessed to obtain the historical operation behavior data.

[0013] In one embodiment, determining the objective function based on the preference probabilities and the costs of the equity includes:

[0014] Will As the objective function, n represents the total number of users, m represents the total number of equity, and P i,j represents the preference probability of the i-th user for the j-th benefit, Uj represents the distribution utility value of the jth equity, C j represents the distribution cost of the j-th stake, and λ represents the weight parameter.

[0015] In one embodiment, the constraint condition includes at least one of the following conditions:

[0016] For each of the users, the quantity of the to-be-distributed benefits distributed does not exceed a first preset quantity threshold;

[0017] For each of the users, the distribution cost of the rights to be distributed does not exceed a preset cost threshold;

[0018] For each of the rights to be distributed, the number of users to whom the rights to be distributed is distributed does not exceed a second preset number threshold.

[0019] In one embodiment, determining the rights to be distributed to each user from the rights database according to the objective function and the preset constraints includes:

[0020] Calculating the matching degree between each of the users and each of the rights and interests;

[0021] Determine a candidate rights and interests set for each of the users from the rights and interests database according to the matching degree;

[0022] The rights to be distributed to each of the users are determined from each of the candidate rights sets according to the objective function and preset constraints.

[0023] In one embodiment, the calculating the matching degree between each user and each benefit includes:

[0024] Obtaining equity characteristic data of each of the equity;

[0025] Calculating the matching degree between each of the users and each of the rights and interests according to the user characteristic data of each of the users and the rights and interests characteristic data;

[0026] Determining a candidate rights set for each user from the rights database according to the matching degree includes:

[0027] For each of the users, a set of rights and interests with a matching degree greater than a preset matching degree threshold is taken as a candidate rights and interests set for the user.

[0028] In one embodiment, distributing the rights to be distributed to the corresponding users includes:

[0029] Selecting a corresponding rights distribution channel for each of the rights to be distributed of each of the users according to the user characteristic data of each of the users and / or the type of each of the corresponding rights to be distributed;

[0030] The rights to be distributed are distributed to the corresponding users through the rights distribution channels.

[0031] In one embodiment, after distributing each of the to-be-distributed benefits to the corresponding users, the method includes:

[0032] Acquire equity distribution effect data, and optimize at least one of the user feature prediction model, the objective function, and the constraint condition based on the equity distribution effect data;

[0033] and / or,

[0034] Acquire feedback data from each of the users regarding the equity distribution result, and optimize at least one of the user feature prediction model, the objective function, and the constraint condition based on the feedback data.

[0035] In another aspect, a rights distribution device is provided, comprising:

[0036] The acquisition module is used to obtain the historical operation behavior data of each user on each right in the current right database;

[0037] A prediction module, used for inputting each of the historical operation behavior data into a trained user feature prediction model to obtain user feature data of each of the users; the user feature data includes the preference probability of the corresponding user for each of the rights and interests;

[0038] A first determination module, configured to determine an objective function based on each of the preference probabilities and the distribution cost of each of the equity;

[0039] A second determination module, configured to determine the rights to be distributed to each user from the rights database according to the objective function and preset constraints;

[0040] The distribution module is used to distribute the rights to be distributed to the corresponding users.

[0041] On the other hand, an electronic device is provided, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to implement any of the above methods.

[0042] Through the rights distribution method, device and electronic device provided by the present application, the historical operation behavior data of each user on each right in the current rights database can be obtained, and the historical operation behavior data can be input into the trained user feature prediction model to obtain the user feature data of each user. The user feature data includes the preference probability of the corresponding user for each right, and the objective function is determined based on each preference probability and the cost of each right. According to the objective function and preset constraints, the rights to be distributed of each user are determined from the rights database, and each right to be distributed is distributed to the corresponding users, thereby realizing the automatic distribution of user rights, allocating rights to each user based on the user's preference probability for each right, and improving the adaptability of rights distribution. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flowchart of the equity distribution method provided in Example 1 of the present application;

[0044] Figure 2 A schematic diagram of the structure of the equity distribution device provided in the second embodiment of the present application;

[0045] Figure 3 This is a schematic diagram of the structure of an electronic device provided in Example 3 of the present application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] Embodiment 1:

[0048] This application embodiment provides a method for distributing rights and interests, which aims to improve the efficiency, security and personalization level of rights and interests distribution through advanced machine learning technology. Figure 1 As shown, the equity distribution method provided in the embodiment of the present application includes the following steps:

[0049] S11: Obtain the historical operation behavior data of each user on each right in the current right database.

[0050] S12: Inputting each historical operation behavior data into the trained user feature prediction model to obtain user feature data of each user, wherein the user feature data includes the preference probability of the corresponding user for each benefit.

[0051] S13: Determine the objective function based on each preference probability and the distribution cost of each equity.

[0052] S14: Determine the rights to be distributed to each user from the rights database according to the objective function and preset constraints.

[0053] S15: Distribute the rights to be distributed to the corresponding users.

[0054] The following is a detailed introduction to the contents of the above steps.

[0055] The equity database in the embodiment of the present application stores a variety of equity, which can be classified and stored according to equity type. The equity stored in the equity database includes but is not limited to at least one of the following equity:

[0056] E-book rights: novels, education, technology, etc.;

[0057] Music rights: pop, classical, rock, etc.;

[0058] Video rights: movies, TV series, documentaries, etc.;

[0059] In-app benefits: virtual currency, game equipment, membership services, etc.

[0060] Each equity in the equity database can have detailed attribute description information, including but not limited to at least one of the following descriptions:

[0061] Name: the name or title of the interest;

[0062] Category: the classification to which the equity belongs;

[0063] Price: the price or value of an interest;

[0064] Description: A detailed description of the rights, such as the content, author, director, etc.

[0065] Tags: Characteristic tags of rights, such as "Popular", "Newly Listed", etc.

[0066] Each equity in the equity database can have corresponding equity evaluation content, where the evaluation indicators include but are not limited to at least one of the following indicators:

[0067] Popularity: popularity based on user clicks, purchases, etc.

[0068] User rating: the user's overall rating of the benefit;

[0069] Expert Rating: Professional rating of the equity by experts or editors.

[0070] The equity database can be updated regularly to introduce new equity and eliminate unpopular equity. The update strategy includes but is not limited to at least one of the following strategies:

[0071] New benefits are launched: new benefits are introduced regularly, such as new books, new movies, etc.

[0072] Delisting of old benefits: Regularly eliminate benefits with low user attention;

[0073] Benefits recommendation: Recommend popular benefits based on user feedback and market trends.

[0074] In step S11, the original historical operation behavior data of each user on each right in the current right database may be collected, and then each original historical operation behavior data may be preprocessed to obtain the historical operation behavior data.

[0075] The original historical operation behavior data in the embodiments of the present application include but are not limited to transaction data, browsing data, social network behavior data, user feedback data, etc. of each user for each benefit. Data collection complies with relevant regulations on privacy protection and data security.

[0076] Exemplarily, the original historical operation behavior data may include:

[0077] Transaction data: user’s purchase history, transaction frequency, preferred payment methods, etc.

[0078] Browsing data: users’ browsing behavior on the platform, including browsed products, dwell time, click-through rate, etc.

[0079] User feedback data: user evaluation, ratings, comments, etc. on products or services.

[0080] The preprocessing steps in the embodiments of the present application include but are not limited to cleaning, noise removal, missing value filling, standardization and feature extraction. Preprocessing the original historical operation behavior data can improve data quality to improve the accuracy of equity distribution.

[0081] Data cleaning refers to removing invalid, erroneous, and incomplete data records. Data denoising refers to using statistical methods to identify and process outliers and noise data to improve data quality. Data standardization refers to normalizing data so that it has the same dimension and value range for subsequent processing.

[0082] It should be noted that in the embodiment of the present application, the above steps can be performed periodically, that is, the rights and interests can be distributed to each user periodically.

[0083] The user feature prediction model in step S12 is a model obtained through training. When training the user feature prediction model, a deep learning model can be selected as the initial model, and the user feature prediction model can be trained using the initial model and the training data set obtained in advance. Among them, a convolutional neural network model or a recurrent neural network model can be selected as the initial model according to business needs. Specifically, for data with a grid structure, such as images, a convolutional neural network model can be selected as the initial model, and for sequence data, such as time series or text data, a recurrent neural network model can be selected as the initial model. The data in the training data set are all pre-processed data, and the specific training process may include:

[0084] Forward propagation: calculate model output;

[0085] Loss function: evaluates the difference between the model output and the true value;

[0086] Backpropagation: Adjust model parameters based on the loss function.

[0087] Model evaluation: Use the validation set to evaluate the performance of the model and select the best model for parameter tuning.

[0088] Assuming that U is the historical operation behavior data and P is the user feature data, the user feature prediction model M can be expressed as P=M(U;θ), where θ represents the model parameters.

[0089] In the embodiment of the present application, a rights distribution strategy can be generated based on the predicted user feature data. Step S13 to step S15 is the process of generating a rights distribution strategy and distributing corresponding rights to each user according to the rights distribution strategy.

[0090] In step S13, the objective function can be determined based on the multi-objective optimization problem and expressed as: Among them, Utility represents the utility function of the equity distribution strategy, Cost represents the cost function of the equity distribution strategy, and λ represents the parameter for adjusting the weight between the utility function and the cost function.

[0091] In order to generate a rights distribution strategy, it is necessary to select a suitable optimization algorithm. In the embodiment of the present application, the following optimization algorithm can be used to generate a rights distribution strategy:

[0092] Linear programming: Applicable to situations where both the objective function and the constraints are linear.

[0093] Dynamic programming: Applicable to multi-stage decision-making problems.

[0094] Genetic Algorithm: Suitable for complex optimization problems, especially those that are difficult to solve with traditional mathematical methods.

[0095] Step S14 in the embodiment of the present application may include the following sub-steps:

[0096] Sub-step 1: Calculate the matching degree between each user and each benefit.

[0097] Sub-step 2: Determine the candidate rights set of each user from the rights database according to the matching degree.

[0098] Sub-step three: Determine the rights to be distributed to each user from each candidate rights set based on the objective function and preset constraints.

[0099] In the above sub-step 1, the equity characteristic data of each equity may be obtained; and the matching degree between each user and each equity may be calculated based on the user characteristic data of each user and the characteristic data of each equity.

[0100] It is understandable that the characteristics of each right can be extracted from the rights database to obtain the rights characteristic data of each right. For example, the rights characteristic data can be determined according to the category of each right, and / or attribute description information, and / or rights evaluation content.

[0101] In addition to the user's preference probability for each benefit, user characteristic data may also include the user's interests, consumption capacity, etc.

[0102] In the embodiment of the present application, the matching degree between each user and each benefit can be calculated in the following way:

[0103] Cosine Similarity: Calculate the cosine similarity between user feature data and equity feature data.

[0104] Euclidean distance: Calculates the Euclidean distance between user feature data and equity feature data.

[0105] Mahalanobis distance: Considering the correlation between features, the Mahalanobis distance between user feature data and equity feature data is calculated.

[0106] In the above sub-step 2, for each user, a set of rights and interests with a matching degree greater than a preset matching degree threshold may be used as a candidate rights and interests set for the user.

[0107] It should be noted that the above sub-step three is equivalent to finding the optimal solution of the objective function, and the constraints should be met in the process of finding the optimal solution of the objective function. The objective function in the embodiment of the present application can be a function determined based on multiple objectives to be optimized, and the objectives to be optimized can include: maximizing user satisfaction, minimizing equity distribution costs, etc. If there are multiple objectives to be optimized, multi-objective optimization techniques, such as Pareto optimization, can be used to find the optimal solution.

[0108] In the embodiment of the present application, As the objective function, n represents the total number of users, m represents the total number of equity, and P i,j represents the preference probability of the i-th user for the j-th benefit, U j represents the distribution utility value of the jth equity, C j represents the distribution cost of the jth equity, and λ represents the weight parameter. The objective to be optimized in this function is to maximize user satisfaction and minimize the equity distribution cost.

[0109] It should be noted that, in the embodiment of the present application, the distribution utility value of each equity can be calculated by the following formula:

[0110] U=w1S+w2E+w3T+W4P;

[0111] Among them, U represents the distribution utility value, S represents the score of user satisfaction related factors, E represents the score of the equity value factor itself, T represents the score of market trend and platform strategy factors, P represents the probability of user preference for the equity, w1, w2, w3, and w4 are the corresponding weight coefficients, and w1+w2+w3+w4=1.

[0112] The user satisfaction related factor score S can be calculated using the formula S=(S1+S2) / 2.

[0113] Historical behavior feedback score S1: It can be quantified based on the user's historical usage frequency, duration, and evaluation of the benefit. For example, for a benefit with high frequency of use, long duration, and high evaluation, S1 can be set to 80-100 points; for a benefit with medium frequency of use, average duration, and medium evaluation, S1 can be set to 50-79 points; for a benefit with low frequency of use, short duration, and low evaluation, S1 can be set to 0-49 points.

[0114] Expected score for improving user activity S2: If the benefit is expected to significantly improve user activity, S2 can be set to 70-100 points; if it has a certain improvement effect, S2 is set to 40-69 points; if the improvement effect is not obvious, S2 is set to 0-39 points.

[0115] S=(S1+S2) / 2.

[0116] The equity intrinsic value factor score E can be calculated using the formula E=(E1+E2) / 2.

[0117] Equity popularity score E1: It can be calculated based on the number of clicks, purchases, favorites and other indicators of the equity. For example, for popular equity (ranked in the top 10% in terms of clicks, purchases and favorites), E1 is set at 80-100 points; for relatively popular equity (ranked in the top 11%-30%), E1 is set at 60-79 points; for general equity (ranked in the top 31%-70%), E1 is set at 40-59 points; for unpopular equity (ranked in the bottom 30%), E1 is set at 0-39 points.

[0118] Expert or professional scoring E2: Different level ranges are set according to the expert scoring. For example, if the expert scoring is 9-10 points, E2 is set to 80-100 points; if the expert scoring is 7-8 points, E2 is set to 60-79 points; if the expert scoring is 5-6 points, E2 is set to 40-59 points; if the expert scoring is 3-4 points, E2 is set to 20-39 points; if the expert scoring is 1-2 points, E2 is set to 0-19 points.

[0119] The market trend and platform strategy factor score T can be calculated using the formula T=(T1+T2) / 2.

[0120] Score T1 for compliance with market trends: If the equity fully complies with the current market trend, T1 is set at 70-100 points; if it is relatively consistent, T1 is set at 40-69 points; if it is not quite consistent but has certain potential, T1 is set at 20-39 points; if it is completely inconsistent, T1 is set at 0-19 points.

[0121] Platform promotion focus score T2: If the rights and interests are currently the focus of the platform's promotion, T2 is set to 70-100 points; if there is a certain degree of promotion, T2 is set to 40-69 points; if the promotion intensity is relatively small, T2 is set to 20-39 points; if not promoted, T2 is set to 0-19 points.

[0122] In the embodiment of the present application, the distribution cost of each equity can be calculated by the following formula:

[0123] C=C1+C2+C3+C4

[0124] Among them, C represents the distribution cost of rights and interests, C1 represents the direct purchase cost, C2 represents the cooperation cost, C3 represents the technical support cost, and C4 represents the promotion and marketing cost.

[0125] Direct purchase cost C1:

[0126] If it is the rights of purchasing copyright, it can be calculated based on the copyright fee. For example, the copyright fee of an e-book is X yuan, then C1 = X.

[0127] If there is a discount, it can be calculated based on the discounted price, that is, C1=X*discount rate.

[0128] Cooperation cost C2:

[0129] The cost of cooperation includes fixed cost F and variable cost V. If you cooperate with game developers to launch game equipment rights, the fixed cost is F yuan, and the variable cost is divided according to user usage or sales. If the sharing ratio is r, and the user usage or sales is Y, then C2 = F + r * Y.

[0130] Technical support cost C3:

[0131] The technical support cost can be apportioned according to the amount of equity distribution. Assume that the total technical support cost including server maintenance fee, data transmission fee is Z yuan, and the amount of equity distribution is N, then C3 = Z / N.

[0132] Promotion and marketing costs C4:

[0133] The promotion and marketing costs can be calculated based on the promotion strength and expected effect of the rights. For example, if the advertising cost is A yuan and the social media promotion cost is B yuan, then C4 = A + B.

[0134] The constraint conditions in the embodiments of the present application may include at least one of the following conditions:

[0135] For each user, the number of rights to be distributed does not exceed a first preset number threshold;

[0136] For each user, the distribution cost of the rights to be distributed does not exceed the preset cost threshold;

[0137] For each benefit to be distributed, the number of users to whom the benefit to be distributed is distributed does not exceed a second preset number threshold.

[0138] The optimization algorithm is used to solve the objective function and constraint conditions to obtain the rights to be distributed to each user.

[0139] It should be noted that in the embodiment of the present application, the first preset quantity threshold, the second quantity threshold and the preset cost threshold can all be flexibly set by the developer according to actual conditions.

[0140] It is understandable that, in some embodiments, a preset cost threshold corresponding to each equity to be distributed may be preset, and in the process of executing step S14, it should be ensured that the distribution cost of each equity to be distributed does not exceed its corresponding preset cost threshold. In other embodiments, a total preset cost threshold may be preset for all equity to be distributed, and in the process of executing step S14, it is ensured that the total distribution cost of each equity to be distributed does not exceed the total preset cost threshold.

[0141] In step S15, corresponding rights distribution channels can be selected for the rights to be distributed of each user according to the user characteristic data of each user and / or the type of the corresponding rights to be distributed; and the rights to be distributed are distributed to the corresponding users through the rights distribution channels.

[0142] The user characteristic data in the embodiment of the present application reflects the characteristics of the user. Therefore, a corresponding rights distribution channel can be selected for the rights to be distributed to the user according to the user characteristic data of the user.

[0143] Exemplarily, the correspondence between age stages and rights distribution channels can be pre-set. In actual applications, the age stage of each user can be determined based on the user characteristic data of each user, and the corresponding rights distribution channel can be determined based on the age stage of each user and the correspondence. Then, the rights to be distributed to the user can be distributed through the rights distribution channel.

[0144] For example, the APP that each user prefers the most can also be determined, and the rights to be distributed to the user can be distributed according to the APP that each user prefers the most. For example, for a user who prefers email, the corresponding rights can be distributed by email.

[0145] Exemplarily, the correspondence between equity types and equity distribution channels can be preset. In actual applications, the corresponding equity distribution channel can be determined according to the type of each equity to be distributed and the correspondence, and then the equity to be distributed can be distributed through the equity distribution channel.

[0146] In the embodiment of the present application, an automated distribution process can also be designed to ensure that the rights and interests can be delivered to users accurately and timely. The distribution process may include:

[0147] Equity packaging: Packaging the equity to be distributed into a format that can be recognized by users.

[0148] Channel adaptation: Adapt the display method of rights and interests according to different distribution channels.

[0149] Message push: Push benefits to users through selected channels.

[0150] In the embodiments of this application, multiple rights distribution channels are used to adapt to different user groups and scenarios:

[0151] Email: Suitable for benefit distribution that requires document form or long-term archiving.

[0152] Mobile application push: suitable for real-time and highly interactive benefits distribution, such as coupons and promotions.

[0153] Social media: using users’ social networks to share and spread benefits.

[0154] After step S15, the following steps may also be included:

[0155] Acquire equity distribution effect data, and optimize at least one of the user feature prediction model, the objective function, and the constraint condition based on the equity distribution effect data;

[0156] and / or,

[0157] Acquire feedback data from each of the users regarding the equity distribution result, and optimize at least one of the user feature prediction model, the objective function, and the constraint condition based on the feedback data.

[0158] The equity distribution effect data in the embodiment of the present application refers to data that can reflect the equity distribution effect, including but not limited to at least one of the following data:

[0159] Delivery rate: the proportion of rights successfully delivered to users;

[0160] Open rate: the proportion of users opening benefits;

[0161] Conversion rate: the proportion of users who convert (such as purchases, registrations) after using benefits.

[0162] In the embodiment of the present application, the feedback data of the user regarding the equity distribution result includes but is not limited to at least one of the following data:

[0163] Satisfaction data: user satisfaction with benefits as reported through satisfaction surveys;

[0164] Usage data: details of the user's use of the rights, such as frequency of use, duration of use, etc.;

[0165] Improvement suggestion data: users’ suggestions for improving the equity distribution system.

[0166] In the embodiment of the present application, at least one of the user feature prediction model, the objective function and the constraint condition can be optimized based on the equity distribution effect data and / or the user's feedback data on the equity distribution result.

[0167] Specifically, the optimization direction of the user feature prediction model includes but is not limited to at least one of the following:

[0168] Feature engineering: adjusting or adding new features to improve the model’s predictive accuracy;

[0169] Parameter tuning: adjust the model's hyperparameters, such as learning rate, regularization coefficient, etc.

[0170] Model structure adjustment: adjust the number of layers and neurons of the model to improve the generalization ability of the model.

[0171] When optimizing the objective function, any parameter in the objective function can be optimized, for example, adjusting the weight parameters mentioned above.

[0172] When optimizing the constraint conditions, any parameter threshold in the optimization condition can be optimized, for example, adjusting the preset cost threshold mentioned above.

[0173] In some embodiments, the equity combination can also be optimized, for example, the equity combination can be adjusted according to user feedback to improve user satisfaction. In some embodiments, the equity distribution channel can also be optimized, for example, the priority of the equity distribution channel can be adjusted according to the delivery rate and the opening rate.

[0174] In the embodiment of the present application, a closed-loop feedback mechanism is constructed to use the optimized user feature prediction model, objective function, and constraint conditions for the next round of equity distribution, forming a continuous optimization cycle.

[0175] In order to ensure the security of rights and interests distribution, in the embodiment of the present application, an encryption algorithm can be used to encrypt the rights and interests to be distributed, and the encrypted rights and interests to be distributed can be sent to the corresponding users.

[0176] In the embodiments of the present application, during the data collection, transmission and storage process, a strong encryption algorithm can be used to encrypt the data to ensure data security. A multi-factor authentication mechanism can be implemented, such as a password + SMS verification code, to ensure that only legitimate users can access the system. According to the user's role and authority, a strict access control policy is implemented to prevent unauthorized access. User data is anonymized to avoid leaking user privacy. In addition, security audits can be conducted regularly to check the security of the rights and interests distribution system, and security vulnerabilities can be discovered and repaired in a timely manner. The response time of the rights and interests distribution system is optimized to ensure that the rights and interests can be delivered to users quickly. Through load balancing, distributed computing and other technologies, the throughput of the rights and interests distribution system is improved to ensure stable operation of the system in high-concurrency scenarios. Optimize the use of rights and interests distribution system resources, such as CPU, memory, storage, etc., to improve resource utilization. Design a fault-tolerant mechanism to ensure that the rights and interests distribution system can recover quickly when a failure occurs.

[0177] The rights distribution method provided in the embodiment of the present application analyzes user behavior through a deep learning algorithm, can more accurately predict user preferences, can improve the efficiency of rights distribution, and utilizes the user behavior analysis model to customize a personalized rights distribution strategy for each user, thereby improving the user experience. It adopts encryption technology and a real-time monitoring feedback mechanism to enhance the security of digital rights distribution, effectively prevent the theft and abuse of rights, monitor user feedback and market changes in real time, and dynamically adjust the distribution strategy to adapt to the ever-changing user needs and market environment.

[0178] It should be understood that, although the various steps in the above-mentioned flow chart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the above-mentioned flow chart may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0179] Embodiment 2:

[0180] See also Figure 2 As shown, the embodiment of the present application also provides a rights distribution device, including:

[0181] Acquisition module 201, used to acquire historical operation behavior data of each user on each right in the current right database;

[0182] Prediction module 202, used for inputting each of the historical operation behavior data into the trained user feature prediction model to obtain user feature data of each of the users; the user feature data includes the preference probability of the corresponding user for each of the rights and interests;

[0183] A first determination module 203, configured to determine an objective function based on each of the preference probabilities and the cost of each of the equity;

[0184] A second determination module 204 is used to determine the rights to be distributed to each user from the rights database according to the objective function and preset constraints;

[0185] The distribution module 205 is used to distribute the rights to be distributed to the corresponding users.

[0186] It should be noted that, for the sake of brevity, the contents described in the above embodiments will not be repeated in this embodiment.

[0187] Embodiment three:

[0188] Based on the same inventive concept, the present application embodiment provides an electronic device, see Figure 3 As shown, it includes a processor 301 and a memory 302. The memory 302 stores a computer program, the processor 301 and the memory 302 communicate via a communication bus, and the processor 301 executes the computer program to implement the steps of the method described above, which will not be repeated here.

[0189] It should be noted that the electronic device in the embodiment of the present application can be a terminal or a server. Among them, the terminal can be a smart phone, a tablet computer, a laptop, a personal computer (PC), a smart home, a wearable electronic device, a VR / AR device, a car computer, etc. The server can be an intercommunication server or a background server between multiple heterogeneous systems, or an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, and basic cloud computing services such as big data and artificial intelligence platforms, etc.

[0190] The processor 301 may be an integrated circuit chip with signal processing capabilities. The processor 301 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. It may implement or execute various methods, steps and logic block diagrams disclosed in the embodiments of the present application. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0191] The memory 302 may include, but is not limited to, random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable read-only memory (EPROM), electrically erasable read-only memory (EEPROM), etc.

[0192] This embodiment also provides a computer-readable storage medium, such as a floppy disk, a CD, a hard disk, a flash memory, a USB flash disk, an SD card, an MMC card, etc., in which one or more programs for implementing the above steps are stored. These one or more programs can be executed by one or more processors to implement the steps of the method in the above embodiment 1, which will not be repeated here.

[0193] It should be noted that the diagram provided in the present embodiment only illustrates the basic concept of the present invention in a schematic manner, so the diagram only shows the components related to the present invention rather than drawing according to the number, shape and size of the components during actual implementation. The type, quantity and ratio of each component during actual implementation can be a random change, and the component layout type may also be more complicated. The structure, ratio, size, etc. illustrated in the drawings of the present specification are only used to match the content disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the limiting conditions that the present invention can implement, so they have no technical substantive significance. Any modification of the structure, change of the proportional relationship or adjustment of the size should still fall within the scope of the technical content disclosed by the present invention without affecting the effect that the present invention can produce and the purpose that can be achieved. At the same time, the terms such as "upper", "lower", "left", "right", "middle" and "one" quoted in this specification are only for the convenience of narration, and are not used to limit the scope of the present invention. The change or adjustment of its relative relationship should also be regarded as the scope of the present invention without substantially changing the technical content.

[0194] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0195] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for distributing equity, characterized in that: include: Obtain the historical operation behavior data of each user on each equity in the current equity database; Inputting each of the historical operation behavior data into the trained user feature prediction model to obtain user feature data of each of the users; the user feature data includes the preference probability of the corresponding user for each of the rights and interests; Determining an objective function based on each of the preference probabilities and the distribution cost of each of the benefits; Determining the rights and interests to be distributed to each of the users from the rights and interests database according to the objective function and preset constraints; Distribute the rights to be distributed to the corresponding users.

2. The equity distribution method according to claim 1, characterized in that: The acquisition of historical operation behavior data of each user on each right in the current right database includes: Collect the original historical operation behavior data of each user on each right in the current right database; The original historical operation behavior data are preprocessed to obtain the historical operation behavior data.

3. The equity distribution method according to claim 1, characterized in that: The determining of the objective function based on the preference probabilities and the costs of the equity includes: Will As the objective function, n represents the total number of users, m represents the total number of equity, and P i,j represents the preference probability of the i-th user for the j-th benefit, U j represents the distribution utility value of the jth equity, C j represents the distribution cost of the j-th stake, and λ represents the weight parameter.

4. The equity distribution method according to claim 1, characterized in that: The constraint condition includes at least one of the following conditions: For each of the users, the quantity of the to-be-distributed benefits distributed does not exceed a first preset quantity threshold; For each of the users, the distribution cost of the rights to be distributed does not exceed a preset cost threshold; For each of the rights to be distributed, the number of users to whom the rights to be distributed is distributed does not exceed a second preset number threshold.

5. The equity distribution method according to claim 1, characterized in that: Determining the rights and interests to be distributed to each user from the rights and interests database according to the objective function and the preset constraint conditions includes: Calculating the matching degree between each of the users and each of the rights and interests; Determine a candidate rights and interests set for each of the users from the rights and interests database according to the matching degree; The rights to be distributed to each of the users are determined from each of the candidate rights sets according to the objective function and preset constraints.

6. The equity distribution method according to claim 5, characterized in that: The calculating the matching degree between each user and each interest includes: Obtaining equity characteristic data of each of the equity; Calculating the matching degree between each of the users and each of the rights and interests according to the user characteristic data of each of the users and the rights and interests characteristic data; Determining a candidate rights set for each user from the rights database according to the matching degree includes: For each of the users, a set of rights and interests with a matching degree greater than a preset matching degree threshold is taken as a candidate rights and interests set for the user.

7. The equity distribution method according to claim 1, characterized in that: The distributing the rights to be distributed to the corresponding users includes: Selecting a corresponding rights distribution channel for each of the rights to be distributed of each of the users according to the user characteristic data of each of the users and / or the type of each of the corresponding rights to be distributed; The rights to be distributed are distributed to the corresponding users through the rights distribution channels.

8. The equity distribution method according to claim 1, characterized in that: After distributing the rights to be distributed to the corresponding users, the method includes: Acquire equity distribution effect data, and optimize at least one of the user feature prediction model, the objective function, and the constraint condition based on the equity distribution effect data; and / or, Acquire feedback data from each of the users regarding the equity distribution result, and optimize at least one of the user feature prediction model, the objective function, and the constraint condition based on the feedback data.

9. A rights distribution device, characterized in that: include: The acquisition module is used to obtain the historical operation behavior data of each user on each right in the current right database; A prediction module, used for inputting each of the historical operation behavior data into a trained user feature prediction model to obtain user feature data of each of the users; the user feature data includes the preference probability of the corresponding user for each of the rights and interests; A first determination module, configured to determine an objective function based on the preference probabilities and the costs of the equity; A second determination module, configured to determine the rights to be distributed to each user from the rights database according to the objective function and preset constraints; The distribution module is used to distribute the rights to be distributed to the corresponding users.

10. An electronic device, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and the processor executes the computer program to implement the method according to any one of claims 1 to 8.

Images