Virtual resource allocation method and device, electronic equipment and storage medium

Abstract

The invention discloses a virtual resource allocation method and device, electronic equipment and a storage medium, and is applied to the field of computers. According to the method, the preference degree of a corresponding object to a target media object is determined according to basic feature data of the target object. And in combination with the preference degree, two absolute earnings generated by the target media object for the object under two conditions of virtual resource allocation and no virtual resource allocation are respectively determined, and the two absolute earnings are respectively embodied through a first allocation index and a second allocation index. According to the difference between the first allocation index and the second allocation index, a virtual resource allocation indication parameter corresponding to the target object is determined, the virtual resource allocation indication parameter can be actually understood as a relative income calculated according to the two absolute incomes, and the relative income is used as a basis for virtual resource allocation. The interaction effect of media objects can be improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and particularly to a virtual resource allocation method, apparatus, electronic device, and storage medium. Background Art

[0002] With the continuous development and popularization of Internet technology, people's demands for information interaction are also constantly changing. Many application programs promote information interaction by presenting media objects to objects, increasing object stickiness. To enhance the interaction effect, virtual resource allocation technology can be used to promote the occurrence of interaction behaviors related to the media object, such as promoting the incidence of exposure click behaviors, click conversion behaviors, and exposure conversion behaviors related to the media object.

[0003] However, the improvement effect or promotion effect generated by the virtual resource allocation scheme in the related art fails to meet expectations, and there is a large room for improvement. Summary of the Invention

[0004] An embodiment of this application provides a virtual resource allocation method, which solves at least one of the foregoing technical problems.

[0005] According to one aspect of the embodiments of this application, a virtual resource allocation method is provided. The method includes:

[0006] Obtain basic feature data corresponding to a target object;

[0007] Determine an object preference parameter corresponding to the target object according to the basic feature data corresponding to the target object, where the object preference parameter indicates the preference degree of the target object for a target media object;

[0008] Respectively determine a first allocation index and a second allocation index corresponding to the target object according to the object preference parameter corresponding to the target object. The first allocation index indicates the probability that the target object performs a preset interaction operation on the target media object when virtual resources are allocated, and the second allocation index indicates the probability that the target object performs a preset interaction operation on the target media object when virtual resources are not allocated; determine a virtual resource allocation indication parameter of the target object for the target media object according to the difference between the first allocation index and the second allocation index;

[0009] Perform virtual resource allocation according to the virtual resource allocation indication parameter of the target object for the target media object.

[0010] According to one aspect of the embodiments of this application, a virtual resource allocation apparatus is provided. The apparatus includes:

[0011] A data acquisition module for acquiring basic feature data corresponding to a target object;

[0012] A virtual resource allocation module for performing the following operations:

[0013] Based on the basic feature data corresponding to the target object, determine the object preference parameter corresponding to the target object, where the object preference parameter indicates the preference degree of the target object for the target media object; based on the object preference parameter corresponding to the target object, respectively determine the first allocation index and the second allocation index corresponding to the target object, where the first allocation index indicates the probability that the target object performs a preset interaction operation on the target media object when virtual resources are allocated, and the second allocation index indicates the probability that the target object performs a preset interaction operation on the target media object when virtual resources are not allocated; based on the difference between the first allocation index and the second allocation index, determine the virtual resource allocation indication parameter of the target object for the target media object; based on the virtual resource allocation indication parameter of the target object for the target media object, perform virtual resource allocation.

[0014] According to one aspect of the embodiments of the present application, a computer device is provided, where the computer device includes a processor and a memory, and at least one instruction, at least one program, a code set or an instruction set is stored in the memory, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the above virtual resource allocation method.

[0015] According to one aspect of the embodiments of the present application, a computer-readable storage medium is provided, where at least one instruction, at least one program, a code set or an instruction set is stored in the storage medium, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the above virtual resource allocation method.

[0016] According to one aspect of the embodiments of the present application, a computer program product is provided, where the computer program product includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions so that the computer device executes to implement the above virtual resource allocation method.

[0017] The technical solutions provided by the embodiments of the present application can bring the following beneficial effects:

[0018] An embodiment of the present application provides a virtual resource allocation method. This virtual resource allocation method determines the preference degree of an object itself for a target media object based on the basic feature data corresponding to the object (target object). Combining this preference degree, two absolute benefits generated by the target media object for the object are respectively determined in two cases: when virtual resources are allocated and when virtual resources are not allocated. These two absolute benefits are respectively reflected by a first allocation index and a second allocation index. The first allocation index indicates the probability that the target object performs a preset interaction operation on the target media object in the case of allocating virtual resources, and the second allocation index indicates the probability that the target object performs a preset interaction operation on the target media object in the case of not allocating virtual resources. According to the difference between the first allocation index and the second allocation index, a virtual resource allocation indication parameter corresponding to the target object is determined. This virtual resource allocation indication parameter can actually be understood as a relative benefit calculated based on the aforementioned two absolute benefits. Using this relative benefit as the basis for virtual resource allocation can solve the technical problems in the related art that using absolute benefits to allocate virtual resources may result in the interaction improvement effect not meeting expectations, or reducing the upper limit of the improvement of click-through rate or exposure rate. It can significantly improve the interaction effect and return on investment of the target media object. For example, it can improve the click-through rate or exposure rate of the target media object, and can also significantly improve the interaction frequency of the target object and enhance the object viscosity. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 is a schematic diagram of an application program running environment provided by an embodiment of the present application;

[0021] Figure 2 is a flowchart of a virtual resource allocation method provided by an embodiment of the present application;

[0022] Figure 3 is a schematic flowchart of an object preference parameter prediction method provided by an embodiment of the present application;

[0023] Figure 4 is a schematic flowchart of a training method for a logistic regression model provided by an embodiment of the present application;

[0024] Figure 5 is a schematic flowchart of a method for determining a first allocation index and a second allocation index provided by an embodiment of the present application;

[0025] Figure 6 It is a schematic diagram of a virtual resource allocation interface provided by an embodiment of the present application;

[0026] Figure 7 It is a block diagram of a virtual resource allocation device provided by an embodiment of the present application;

[0027] Figure 8 It is a structural block diagram of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0028] Before introducing the method embodiments provided by the present application, relevant terms or nouns that may be involved in the method embodiments of the present application are briefly introduced first, so as to facilitate the understanding of those skilled in the art of the present application.

[0029] Cloud technology refers to a hosting technology that unifies a series of resources such as hardware, software, and networks within a wide area network or a local area network to achieve data computing, storage, processing, and sharing. Cloud technology is the general term for network technology, information technology, integration technology, management platform technology, application technology, etc. based on the cloud computing business model, and can form a resource pool, which can be used on demand and is flexible and convenient. Cloud computing technology will become an important support. The background services of the technical network system require a large amount of computing and storage resources, such as video websites, picture websites, and more portal websites. With the high development and application of the Internet industry, in the future, each item may have its own identification mark and needs to be transmitted to the background system for logical processing. Data at different levels will be processed separately, and various industry data requires a powerful system back-end support, which can only be achieved through cloud computing.

[0030] Artificial Intelligence (AI) is to use a digital computer or a machine controlled by a digital computer to simulate, extend, and expand human intelligence, and is a theory, method, technology, and application system that can perceive the environment, acquire knowledge, and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology of computer science. It attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a way similar to human intelligence. Artificial intelligence also studies the design principles and implementation methods of various intelligent machines, so that the machines have the functions of perception, reasoning, and decision-making.

[0031] Artificial intelligence technology is a comprehensive discipline that involves a wide range of fields, including both hardware-level and software-level technologies. The basic technologies of artificial intelligence generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, and mechatronics. The software technologies of artificial intelligence mainly include several major directions such as computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0032] Machine Learning (ML) is an interdisciplinary subject that involves multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It specifically studies how computers simulate or implement human learning behaviors to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve their own performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent, and its applications cover all fields of artificial intelligence. Machine learning and deep learning usually include technologies such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and rote learning.

[0033] Media object: A media object refers to various forms of information, content, and / or resources presented on various media platforms. Media objects can have many forms of expression.

[0034] Natural exposure of media objects: Different from virtual resource allocation technology, natural exposure does not require additional allocation of virtual resources to promote media objects. Instead, it maintains the original exposure rate or exposure method of the media object without change and keeps the original interaction rules of the media object unchanged. For example, if the original exposure position of the media object is in a regular advertising slot, then its exposure in the regular advertising slot is maintained. If the media object does not require rewarding virtual resources after interaction, then the rule of not rewarding is maintained.

[0035] Before specifically elaborating on the embodiments of this application, the relevant technical background related to the embodiments of this application is introduced to facilitate the understanding of those skilled in the art of this application.

[0036] With the continuous development and popularization of Internet technology, people's needs for information interaction are also constantly changing. Many applications promote information interaction by presenting media objects to objects, increasing object stickiness. To enhance the interaction effect, virtual resource allocation technology can be used to promote interaction behaviors related to the media object, such as promoting the occurrence of exposure click behaviors, click conversion behaviors, and exposure conversion behaviors related to the media object. An exposure click behavior refers to the behavior of triggering a click on a media object when the media object is exposed. The click behavior represents a shallow interaction between an object and the application presenting the media object. The click behavior can trigger the application to display more details or interaction information of the media object, etc. A click conversion behavior refers to the behavior of triggering a conversion on a media object when the media object is clicked. Click conversion represents a deeper interaction between an object and the application presenting the media object. For example, it can obtain the item corresponding to the media object, participate in activities related to the media object, or download the application software corresponding to the media object, register and log in to the application software corresponding to the media object, etc. An exposure conversion behavior refers to the behavior of triggering a conversion on a media object when the media object is exposed. If, for a certain media object, the exposure click behavior and the click conversion behavior occur in sequence, it can be considered that an exposure conversion behavior for the media object has occurred.

[0037] Virtual resources can be resources that can circulate and be used in a specific application. Taking a game application as an example, they can be game props or game coins. Virtual resources can also be resources that can circulate and be used in an alliance application. Taking the e-commerce field as an example, they can be resources that can circulate and be used among multiple e-commerce applications with an alliance relationship. For example, multi-platform universal vouchers. Of course, they can also be resources that can circulate and be used in the broad market. For example, virtual coins that can be cashed out, etc. Virtual resource allocation technology refers to a technical means of promoting a media object by allocating virtual resources to an object or a media object through a virtual resource allocation scheme, so as to enhance the information interaction effect of the media object, and can be understood as a promotion or incentive scheme for the media object. There can be many specific implementation methods for virtual resource allocation technology.

[0038] For example, the object can be guided to participate in the interaction of the media object in the form of virtual resource rewards. Taking the e-commerce scenario as an example, when the object obtains the media object, coupons or vouchers can be given to the object, or virtual currency can be returned after the e-commerce transaction of the media object is successfully completed. Then, the coupons, vouchers or virtual currency, etc. are the allocated virtual resources. For another example, high-quality advertising positions or high-quality exposure opportunities are allocated to the media object. For example, the media object is displayed when the application is opened, or the media object is displayed on the first screen of the application, or the media object is displayed when the object performs a high-frequency operation. The high-quality advertising position and the high-quality exposure opportunity are the allocated virtual resources.

[0039] In the related art, the application can model the click-through rate triggered by the object on the media object as the revenue provided by the media object to the object on the premise that virtual resources can be allocated within a period of time. This revenue, as an absolute revenue, is used as the basis for deciding whether to allocate virtual resources to the object and the media object subsequently in the related art. However, this decision-making method only considers the premise of virtual resource allocation, and does not consider the situation of not allocating virtual resources, or rather does not consider the absolute revenue provided by the media object to the object under natural exposure. The embodiments of the present application point out that such technical solutions may lead to a decision result that cannot achieve better revenue when deciding whether to allocate virtual resources to the object and the media object, thus having an adverse impact on the interaction effect of the media object. For example, the improvement effect of the click-through rate of the media object cannot reach the expected value, or the upper limit of the improvement of the click-through rate of the media object is reduced.

[0040] For example, there are object 1 and object 2. The behavioral results of these two objects under the conditions of virtual resource allocation and non-allocation of virtual resources are statistically analyzed, and the statistical results are shown in Table 1.

[0041] Table 1

[0042]

[0043] According to Table 1, it can be seen that under the premise of allocating virtual resources, the click-through rate of object 1 on product A is 0.8, while the click-through rate of object 2 on product A is 2.00. Therefore, according to the decision-making method of the related art, virtual resources will be allocated to object 2 to improve the click-through rate corresponding to product A. The click-through rate is the probability that the click behavior on product A occurs when the exposure behavior on product A occurs, that is, the probability that the exposure click behavior occurs.

[0044] However, in fact, from the perspective of relative gain, the relative gain of Object 1 is greater. Relative gain represents a relative benefit, which means the difference between the absolute benefit provided by the media object to the object when virtual resources are allocated to the media object and the absolute benefit provided by the media object to the object when the media object is naturally exposed. Since the relative gain of Object 1 is greater, choosing to allocate virtual resources to Object 1 can achieve better benefits, that is, better improve the click-through conversion rate.

[0045] The embodiments of the present application propose that the technical solution in the related art for allocating virtual resources for media content based on absolute benefits such as click-through rate or conversion rate of exposure when virtual resources are allocated to the media object does not take into account the absolute benefits generated when the media object is naturally exposed. Therefore, it is impossible to make a decision on whether to allocate virtual resources from the perspective of relative gain, resulting in the virtual resource allocation result being unable to achieve better benefits, thus having an adverse impact on the interaction effect of the media object. For example, the improvement effect of the allocated virtual resources on the click-through conversion rate or conversion rate of exposure of the media object cannot reach the expected value, or the upper limit of the improvement of the click-through conversion rate or conversion rate of exposure of the media object is reduced.

[0046] In view of this, an embodiment of the present application proposes a virtual resource allocation method. This virtual resource allocation method determines the preference degree of an object (target object) for a target media object according to the basic feature data corresponding to the object itself. Combining this preference degree, two absolute benefits generated by the target media object for the object are respectively determined in two cases: when virtual resources are allocated and when virtual resources are not allocated. These two absolute benefits are respectively reflected by a first allocation index and a second allocation index. The first allocation index indicates the probability that the target object performs a preset interaction operation on the target media object in the case of allocating virtual resources. The second allocation index indicates the probability that the target object performs a preset interaction operation on the target media object in the case of not allocating virtual resources. According to the difference between the first allocation index and the second allocation index, a virtual resource allocation indication parameter corresponding to the target object is determined. This virtual resource allocation indication parameter can actually be understood as a relative benefit calculated based on the aforementioned two absolute benefits. Using this relative benefit as the basis for virtual resource allocation can solve the technical problems in the related art that using absolute benefits to allocate virtual resources results in the inability to achieve the expected interaction improvement effect, or reduces the upper limit of the improvement of click-through rate or exposure rate, and significantly improves the interaction effect and return on investment of the target media object. Return on investment, that is, ROI, is the abbreviation of Return On Investment. It is usually used to measure the ability of a task to generate benefits. The basic calculation formula of ROI = benefit / cost. Generally, it is considered that ROI > 1 indicates a positive direction, that is, the task generates positive benefits and has the value of execution. There are many manifestations of improving the return on investment. For example, improving the click-through rate or exposure rate of the target media object can also significantly improve the interaction frequency of the target object and enhance the object viscosity.

[0047] The principle of using relative benefits as the decision-making basis in the embodiment of the present application can be reflected by the following content:

[0048] Consider a group of people with a number of 2n (n > 1). Among them, n people are assigned to the experimental group (labeled as t), and this experimental group performs a certain specific task on each person (for example, allocating virtual resources). The other n people are assigned to the control group (labeled as c), and this control group does not perform this specific task on anyone. For the sake of simplicity, in the embodiment of the present application, i = 1,..., n is defined as the experimental group, and i = n + 1,..., 2n is defined as the control group. For each object, the original execution result data (absolute benefit data) and the converted relative benefit data are respectively y i and z i , z i is a function of y i as a variable, z iThe specific content can be uniquely determined by the following formula and will not be elaborated here. The calculation method of the relative gain generated by this group of people is as follows:

[0049]

[0050] The relative gain can indicate the relative benefit of performing a certain specific task. That is to say, by directly establishing a regression model based on z, for a certain object characterized by its corresponding feature x, the relative execution benefit or relative execution effect of this certain feature task, that is, the relative gain(x) = E[y|x,t] - E[y|x,c] = E[z|x], which can be used as the theoretical basis for allocating virtual resources in the embodiments of the present application.

[0051] To make the objectives, technical solutions, and advantages of the present application clearer, the following will further describe the embodiments of the present application in detail with reference to the accompanying drawings.

[0052] It should be noted before the description that all the data used in the embodiments of this application have been desensitized data, and are legal use data fully authorized by relevant parties.

[0053] Please refer to Figure 1 , which shows a schematic diagram of an application program running environment provided by an embodiment of the present application. The application program running environment may include: a terminal 10 and a server 20.

[0054] The terminal 10 includes but is not limited to electronic devices such as mobile phones, computers, intelligent voice interaction devices, intelligent home appliances, vehicle-mounted terminals, game consoles, e-book readers, multimedia playback devices, and wearable devices. A client of the application program may be installed in the terminal 10.

[0055] In the embodiments of the present application, the above application program may be any application program that provides virtual resource allocation services. Typically, the application program may be a game application program. Of course, in addition to game application programs, virtual resource allocation services may also be provided in other types of application programs. For example, news application programs, social application programs, interactive entertainment application programs, browser application programs, shopping application programs, content sharing application programs, virtual reality (VR) application programs, augmented reality (AR) application programs, application store application programs, etc., which are not limited in the embodiments of the present application. The embodiments of the present application are not limited thereto. Optionally, the client of the above application program runs in the terminal 10.

[0056] Server 20 is used to provide background services for the clients of the application programs in terminal 10. For example, server 20 can be the background server of the above application program. Server 20 can be a single server or a server cluster composed of multiple physical servers, that is, any server in a distributed cluster. This distributed cluster can concurrently respond to requests from multiple clients 10, and it can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0057] Optionally, terminal 10 and server 20 can communicate with each other through network 30. Terminal 10 and server 20 can be directly or indirectly connected through wired or wireless communication methods, and this application does not limit this.

[0058] Please refer to Figure 2 , which shows a flowchart of a virtual resource allocation method provided by an embodiment of this application. This method can be applied to a computer device, and the above computer device refers to an electronic device with data calculation and processing capabilities. For example, the execution entity of each step can be Figure 1 any terminal 10 or server 20 in the application program running environment shown. This method can include the following steps:

[0059] S201. Obtain the basic feature data corresponding to the target object.

[0060] In the embodiments of this application, an object can interact with a media object in an application program. For example, it can generate exposure click behaviors, click conversion behaviors, and / or exposure conversion behaviors for the media object, that is, generate click-through rates and / or conversion rates. The target object can refer to any object in the application program.

[0061] The embodiments of this application do not limit the media object. For example, it can be a game prop provided in a game application program, or a commodity provided in an e-commerce application program, or audio-visual resources provided by an audio-visual platform.

[0062] In the embodiments of this application, the basic feature data refers to the feature data related to predicting the preference parameters of the object. This application does not limit the content of the basic feature data. For example, it can be object attribute data and / or object target records.

[0063] Object attribute data can be understood as the features that describe the static portrait of the object itself, which is related to the specific application program and refers to the static attribute data used to describe the object in the application program. Object attribute data is the existing data in the application program. Taking a game application program as an example, object attribute data is the static attribute data used to describe the object in the game application program, such as the game characters associated with the object, the game props associated with the object, etc.

[0064] The object target record may include at least one of the following contents:

[0065] In-game interaction features: game login features, login status change features, game online features, online status change features, game character combat power value change features.

[0066] Activity participation features: the number of items collected, the number of item clicks, the number of items obtained, the number of gift packs received at each level, the number of coupons received at each level, the number of tasks completed, the number of gifts given, etc.

[0067] In some embodiments, at least one of the following preprocessings may also be performed on the original basic feature data to obtain the preprocessed basic feature data, and the preprocessed basic feature data can be used as the basic feature data in step S201.

[0068] The embodiments of the present application do not limit the preprocessing method. For example, at least one of the following processing methods can be adopted: missing value processing, feature engineering processing.

[0069] Missing value processing: The general method for dealing with missing values is to delete the missing value records or fill them with specified values. If data integrity is considered, the missing values can be filled with the average value of the data.

[0070] Feature engineering processing includes at least one of the following operations: discretization, normalization.

[0071] Discretization: The purpose of discretization is to increase the generalization ability. For example, the feature of the game character level can be discretized by using the equal-interval method. The specific operation is to group the features according to the same principle of each grouping interval. For example, the value range of the game character level is 1-20, and the equal-interval discretization takes 5 as the interval. The game character level is divided into [1, 5], [6, 10], [11, 15], [16, 20] in turn. The game character levels falling into the same interval are given the same game character level feature, so as to achieve the purpose of discretization.

[0072] Normalization: The normalization calculation formula is as follows. The purpose of normalization is to speed up the output speed of the calculation result.

[0073] x' = (x - X_min) / (X_max - X_min), where x represents the original data, X_min represents the minimum value of the original data, X_max represents the maximum value of the original data, and x' represents the value after normalization.

[0074] S202. Determine the object preference parameter corresponding to the target object according to the basic feature data corresponding to the target object, where the object preference parameter indicates the preference degree of the target object for the target media object.

[0075] In the embodiments of the present application, the target media object can be any kind of media object. For example, a certain game package in a game application, or a certain commodity in an e-commerce program, etc. The embodiments of the present application do not limit the specific method for determining the preference degree of the target object for the target media object, that is, the specific calculation method of the object preference parameter, and relevant technologies can be referred to. For example, in related technologies, an artificial intelligence technology is used to train a model that can predict object preferences, and this model can be used to determine the object preference parameter corresponding to the target object. Of course, in order to reduce the training cost and save machine resources on the basis of ensuring the accuracy of the object preference parameter, the embodiments of the present application provide an object preference parameter prediction method. Please refer to Figure 3 , which shows the schematic flowchart of the object preference parameter prediction method provided by the embodiments of the present application. Determining the object preference parameter corresponding to the target object according to the basic feature data corresponding to the target object includes:

[0076] S301. Extract target feature data from the basic feature data corresponding to the target object, where the target feature data includes at least one of the following types of features: the attribute features corresponding to the target object, and the interaction features generated by the target object performing operations on the target media object.

[0077] The basic feature data may include interaction features that are irrelevant to the target media object. For example, interaction features generated by behaviors such as clicking, obtaining, using, and liking other media objects other than the target media object. These interaction features have a low correlation with preference prediction, so they can be removed in step S301 to obtain the target feature data. The embodiments of the present application do not limit the specific operation content of the target object performing operations on the target media object. For example, if the target media object is a commodity, the specific operation content can be clicking, viewing, consulting, commenting, obtaining, using, etc.; if the target media object is a short video, the specific operation content can be clicking to play, long playing, collecting, commenting, liking, complaining, following the author, not interested, etc.

[0078] S302. Perform logistic regression processing on the target feature data to obtain the object preference parameter corresponding to the target object.

[0079] The logistic regression processing in the embodiments of this application is implemented based on a logistic regression model. Please refer to Figure 4 , which shows a schematic flowchart of the training method of the logistic regression model in the embodiments of this application. The above logistic regression model is trained through the following method:

[0080] S401. Obtain the sample basic feature data corresponding to the sample object and the preference label of the sample object for the sample media object; extract sample feature data from the above sample basic feature data, and the above sample feature data includes at least one of the following types of features: the attribute features corresponding to the above sample object, and the interaction features generated by the above sample object performing operations on the above sample media object.

[0081] S402. Input the above sample feature data into the logistic regression model, trigger the logistic regression model to calculate a preference mapping value according to the above sample feature data and a preset mapping formula, and perform normalization processing on the above preference mapping value to obtain a preference prediction value, where the above preset mapping formula is an exponential function including trainable parameters.

[0082] In the embodiments of this application, normalization processing is generally used to limit data within a certain range, usually between 0 and 1, or between a minimum value and a maximum value set according to actual needs. There are many specific normalization functions. In the embodiments of this application, the choice of the normalization processing function is not limited. For example, the sigmoid function can be used, and this function can convert the target value into a certain value between 0 and 1. The sigmoid function is a common S-shaped function and will not be elaborated in the embodiments of this application.

[0083] In the embodiments of this application, there is also no limitation on the preset mapping formula. It is an exponential function of trainable parameters and can correctly express the correlation between the sample feature data and the preference mapping value after training. In one embodiment, the preset mapping formula includes where w is the trainable parameter and x is the sample feature data. The above preset mapping formula reflects the combined result of the preference mapping operation and the normalization processing function, that is

[0084] S403. Adjust the value of the above trainable parameter according to the difference between the above preference prediction value and the above preference label until the training stop condition is reached.

[0085] The embodiments of the present application do not limit the quantification method of the difference between the above-mentioned preference prediction value and the above-mentioned preference label. The preference label is a binary label, while the preference prediction value is a probability value. Therefore, the difference between the above-mentioned preference prediction value and the above-mentioned preference label is easily quantifiable. For example, direct subtraction or cross-entropy quantification can be used. The present application also does not limit the adjustment method of the trainable parameters and the training stop condition. For example, when the above-mentioned difference is less than the preset difference threshold or the number of parameter adjustment times is greater than the preset number threshold, the parameter adjustment can be stopped. Of course, the embodiments of the present application do not limit the magnitudes of the preset difference threshold and the preset number threshold.

[0086] S203. According to the object preference parameters corresponding to the above-mentioned target object, respectively determine the first allocation index and the second allocation index corresponding to the above-mentioned target object. The first allocation index indicates the probability that the above-mentioned target object performs a preset interaction operation on the above-mentioned target media object when virtual resources are allocated. The second allocation index indicates the probability that the above-mentioned target object performs a preset interaction operation on the above-mentioned target media object when virtual resources are not allocated. Determine the virtual resource allocation indication parameter of the above-mentioned target object for the above-mentioned target media object according to the difference between the first allocation index and the second allocation index.

[0087] In the embodiments of the present application, the first allocation index and the second allocation index actually represent the absolute benefits in two cases respectively. The former is the absolute benefit when virtual resources are allocated, and the latter is the absolute benefit when virtual resources are not allocated. The determination of the absolute benefit in the embodiments of the present application mainly models the probability of the preset interaction operation, such as the probability of occurrence of exposure click behavior, click conversion behavior, or exposure conversion behavior, etc. That is to say, the absolute benefit modeled in the embodiments of the present application can be reflected by the click-through rate or conversion rate. If the click-through rate or conversion rate is high, it can be determined that the corresponding absolute benefit is high. The virtual resource allocation indication parameter reflects the difference between the first allocation index and the second allocation index, that is, the relative benefit, that is, the additional benefit generated by allocating virtual resources compared to not allocating virtual resources.

[0088] Please refer to Figure 5 , which shows a schematic flowchart of the method for determining the first allocation index and the second allocation index in the embodiments of the present application. The above-mentioned steps of respectively determining the first allocation index and the second allocation index corresponding to the above-mentioned target object according to the object preference parameters corresponding to the above-mentioned target object include:

[0089] S501. Obtain the first reference index corresponding to the above-mentioned target object. The first reference index indicates the probability that the above-mentioned target object performs a preset interaction behavior when virtual resources are allocated. Obtain the second reference index corresponding to the above-mentioned target object. The second reference index indicates the probability that the above-mentioned target object performs a preset interaction behavior when virtual resources are not allocated.

[0090] In the embodiments of the present application, P(Y i |x i ,T i = 1) and P(Y i |x i ,T i = 0) respectively represent the first allocation index and the second allocation index. Y i represents a preset interaction behavior, x i represents the basic feature data of the above target object, P(Y i |x i ) represents the object preference parameter of the above target object, and P(Y i |x i ) is a known quantity and has been obtained previously. T i represents whether to allocate virtual resources, where T i = 1 indicates that virtual resources have been allocated, and T i = 0 indicates that virtual resources have not been allocated.

[0091] P(T i = 1) indicates the first reference index, and P(T i = 0) indicates the second reference index. The former is the probability that the above target object performs the preset interaction behavior in the case of virtual resource allocation obtained by statistically analyzing the historical preset interaction behavior of the target object. The statistical process does not consider the differences in media objects and only quantifies the probability of the preset interaction behavior in the case of virtual resource allocation. The latter is the probability that the above target object performs the preset interaction behavior in the case of no virtual resource allocation obtained by statistically analyzing the historical preset interaction behavior of the target object. The statistical process does not consider the differences in media objects and only quantifies the probability of the preset interaction behavior in the case of no virtual resource allocation. It can be seen that both the first reference index indicated by P(T i = 1) and the second reference index indicated by P(T i = 0) are known quantities.

[0092] S502. Determine the first allocation index corresponding to the above target object according to the first reference index corresponding to the above target object and the preference parameter corresponding to the above target object; determine the second allocation index corresponding to the above target object according to the second reference index corresponding to the above target object and the preference parameter corresponding to the above target object.

[0093] In the embodiments of the present application, determining the first allocation indicator corresponding to the target object according to the first reference indicator corresponding to the target object and the preference parameter corresponding to the target object includes: determining the first allocation indicator corresponding to the target object according to the product of the first reference indicator corresponding to the target object and the preference parameter corresponding to the target object; determining the second allocation indicator corresponding to the target object according to the second reference indicator corresponding to the target object and the preference parameter corresponding to the target object includes: determining the second allocation indicator corresponding to the target object according to the product of the second reference indicator corresponding to the target object and the preference parameter corresponding to the target object. For example, the product of the first reference indicator corresponding to the target object and the preference parameter corresponding to the target object can be directly used as the first allocation indicator corresponding to the target object, or the first allocation indicator can be obtained based on the product and a preset formula, as long as the first allocation indicator has a positive correlation with the product. The method for obtaining the second allocation indicator is the same as that of the first allocation indicator and will not be elaborated. The preset formula can be set according to actual needs and will not be elaborated.

[0094] In the embodiments of the present application, P(Y i |x i ,T i =1) and P(Y i |x i ,T i =0) respectively represent the first allocation indicator and the second allocation indicator, and P(Y i |x i ,T i =1)-P(Y i |x i ,T i =0) is the virtual resource allocation indication parameter of the target object for the target media object.

[0095] S204. Perform virtual resource allocation according to the virtual resource allocation indication parameter of the target object for the target media object.

[0096] The virtual resource allocation indication parameter of the target object for the target media object indicates the relative benefit generated by the target media object for the target object. In the embodiments of the present application, this relative benefit is used as the basis for virtual resource allocation.

[0097] In one embodiment, the above-mentioned target object is any object in the object set. The virtual resource allocation indication parameter for the above-mentioned target media object according to the above-mentioned target object is used for virtual resource allocation, including: determining the virtual resource allocation indication parameters of other objects in the object set for the above-mentioned target media object; determining the objects in the object set whose virtual resource allocation parameters meet the first condition as virtual resource allocation objects; and allocating virtual resources to the above-mentioned virtual resource allocation objects.

[0098] Of course, the embodiments of the present application do not elaborate on the method for determining the virtual resource allocation indication parameters of other objects in the object set for the above-mentioned target media object. The method for determining the virtual resource allocation indication parameters of other objects in the object set for the above-mentioned target media object is based on the same inventive concept as the method for determining the virtual resource allocation indication parameters of the target object for the above-mentioned target media object, and will not be elaborated here.

[0099] In the embodiments of the present application, the target object can be any object in the pre-acquired target object set. The embodiments of the present application do not limit the method for acquiring the target object set. It can be considered that the target object set is an object set including objects that need to allocate virtual resources. The target object set can be set according to the requirements of the application program itself. The embodiments of the present application do not elaborate on this, and for the embodiments of the present application, the target object set is a known piece of information.

[0100] The embodiments of the present application also do not limit the specific method for determining the objects in the object set whose virtual resource allocation parameters meet the first condition as virtual resource allocation objects, or the specific content of the first condition. The first condition can be set according to actual requirements.

[0101] For example, each object in the object set can be arranged in descending order of the above-mentioned virtual resource allocation parameters to obtain an object sequence; the first preset number of objects in the object sequence can be determined as virtual resource allocation objects; or, the objects in the object set whose corresponding virtual resource allocation parameters are greater than the preset allocation threshold can be determined as virtual resource allocation objects. Of course, the embodiments of the present application do not limit the preset number and the preset allocation threshold, which can be determined according to the total amount of virtual resources. This embodiment can allocate virtual resources to objects with relatively higher returns, significantly promoting the improvement of the conversion rate or click-through rate related to the target media object. For example, the above-mentioned allocating virtual resources to the above-mentioned virtual resource allocation objects includes: when presenting the above-mentioned target media object to the above-mentioned virtual resource allocation objects, sending virtual resources to the above-mentioned virtual resource allocation objects. Taking an e-commerce scenario as an example, if the virtual resource is a voucher, virtual currency, etc., it can be used in the interaction of the virtual resource allocation object with the target media content, thereby significantly improving the interaction effect of the target media content.

[0102] In one implementation manner, the above-mentioned target media object is any media object in the media object set. The virtual resource allocation according to the virtual resource allocation indication parameter of the above-mentioned target object for the above-mentioned target media object includes: determining the virtual resource allocation indication parameter of the above-mentioned target object for other media objects in the media object set; determining the media objects whose corresponding virtual resource allocation parameters in the above-mentioned media object set meet the second condition as associated media objects. The above-mentioned associated media objects are media objects associated with virtual resources, and the above-mentioned virtual resources are display resources; displaying the above-mentioned associated media objects to the above-mentioned target object.

[0103] In the embodiments of the present application, the target media object can be any media object in the pre-acquired media object set. The embodiments of the present application do not limit the method for obtaining the media object set. It can be considered that the media object set is a set including media objects for which virtual resources need to be allocated. The media object set can be set according to the needs of the application program itself. The embodiments of the present application do not elaborate on this, and for the embodiments of the present application, the media object set is a known piece of information.

[0104] Of course, the embodiments of the present application do not elaborate on the method for determining the virtual resource allocation indication parameter of the above-mentioned target object for other media objects in the above-mentioned media object set. It is based on the same inventive concept as the method for determining the virtual resource allocation indication parameter of the above-mentioned target object for the above-mentioned target media object, and will not be elaborated here.

[0105] The embodiments of the present application also do not limit the specific method for determining the media objects whose corresponding virtual resource allocation parameters in the above-mentioned media object set meet the second condition as associated media objects, or the specific content of the second condition. The second condition can be set according to actual needs.

[0106] For example, each media object in the above-mentioned media object set can be arranged in descending order of the corresponding virtual resource allocation parameter to obtain a media object sequence; the first preset number of media objects in the above-mentioned media object sequence can be determined as associated media objects; or, the media objects in the above-mentioned media object set whose corresponding virtual resource allocation parameters are greater than the preset allocation threshold can be determined as associated media objects. Of course, the embodiments of the present application do not limit the preset number and the preset allocation threshold, which can be determined according to the total amount of virtual resources.

[0107] This implementation manner can allocate virtual resources to media objects (associated media objects) with relatively high benefits for the target object. For example, the virtual resource can be a high-quality advertisement display position or a display opportunity. Based on this virtual resource, displaying the associated media objects can be more conducive to the target object receiving and reaching the associated media objects, achieving the effect of displaying the top media objects, and promoting the improvement of the conversion rate or click-through rate related to the media objects.

[0108] Please refer to Figure 6 which shows a schematic diagram of the virtual resource allocation interface provided by an embodiment of the present application. Taking a game application as an example, the game application can display associated media objects adapted to the game player on the media content display interface. The associated media objects are all top media objects with relatively high benefits for the game player, which not only improves the conversion rate and click-through rate, but also provides the ability to display personalized media objects. Because for different game players, the displayed associated media objects can be different. The interface of game player 1 shows "high-level" weapons and "healing" medicines. The interface of game player 2 shows "cultivation" spirit stones and healing props.

[0109] To verify the technical effects of the embodiments of the present application, an AB test was conducted for one week. The AB test is a commonly used experimental design method for comparing the effect differences of two or more different treatment methods or strategies on a certain indicator to determine which method is better. In the AB test, the experimental objects are randomly divided into two groups, one group as the control group and the other group as the experimental group, and different treatment methods or strategies are adopted respectively, and a comparison is made on a certain indicator. The control group usually uses the existing treatment method or strategy as a reference, and the experimental group uses the new treatment method or strategy. By comparing the differences between the two groups on the indicator, it can be evaluated whether the effect of the new treatment method or strategy is significant. In the embodiments of the present application, the experimental group and the control group are also divided. The experimental group uses the technical solution of the embodiments of the present application to allocate virtual resources, and the control group uses the old technical solution to allocate virtual resources. The conversion rate of the experimental group is 32% higher than that of the control group, and the per capita acquisition is 29% higher. It can be seen that the technical solution of the embodiments of the present application has an obvious promoting effect on the virtual resource interaction efficiency.

[0110] Please refer to Figure 7 which shows a block diagram of a virtual resource allocation device provided by an embodiment of the present application. The device has the function of implementing the above virtual resource allocation method. The above function can be implemented by hardware or by hardware executing corresponding software. The device can be a computer device or can be set in a computer device. The above device includes:

[0111] A basic data acquisition module 701, configured to acquire basic feature data corresponding to a target object;

[0112] The virtual resource allocation module 702 is configured to determine, according to the basic feature data corresponding to the above target object, the object preference parameter corresponding to the above target object, where the object preference parameter indicates the preference degree of the above target object for the target media object; determine, according to the object preference parameter corresponding to the above target object, the first allocation index and the second allocation index corresponding to the above target object respectively, where the first allocation index indicates the probability that the above target object performs a preset interaction operation on the above target media object in the case of allocating virtual resources, and the second allocation index indicates the probability that the above target object performs a preset interaction operation on the above target media object in the case of not allocating virtual resources; determine the virtual resource allocation indication parameter of the above target object for the above target media object according to the difference between the above first allocation index and the above second allocation index; and perform virtual resource allocation according to the virtual resource allocation indication parameter of the above target object for the above target media object.

[0113] In one embodiment, the above target object is any object in the object set, and the virtual resource allocation module 702 is configured to perform the following operations:

[0114] Determine the virtual resource allocation indication parameters of other objects in the above object set for the above target media object;

[0115] Determine the objects in the above object set whose virtual resource allocation parameters meet the first condition as virtual resource allocation objects;

[0116] Allocate virtual resources to the above virtual resource allocation objects.

[0117] In one embodiment, the above target media object is any media object in the media object set, and the virtual resource allocation module 702 is configured to perform the following operations:

[0118] Determine the virtual resource allocation indication parameters of the above target object for other media objects in the above media object set;

[0119] Determine the media objects in the above media object set whose corresponding virtual resource allocation parameters meet the second condition as associated media objects, where the associated media objects are media objects associated with virtual resources, and the virtual resources are display resources;

[0120] Display the above associated media objects to the above target object.

[0121] In one embodiment, the virtual resource allocation module 702 is configured to perform the following operations:

[0122] Obtain the first reference index corresponding to the above target object, where the first reference index indicates the probability that the above target object performs a preset interaction behavior in the case of allocating virtual resources;

[0123] Obtain a second reference index corresponding to the above target object, where the second reference index indicates the probability that the above target object performs a preset interaction behavior without virtual resource allocation;

[0124] Determine a first allocation index corresponding to the above target object according to the first reference index corresponding to the above target object and the preference parameter corresponding to the above target object;

[0125] Determine a second allocation index corresponding to the above target object according to the second reference index corresponding to the above target object and the preference parameter corresponding to the above target object.

[0126] In one embodiment, the above virtual resource allocation module 702 is used to perform the following operations:

[0127] Determine a first allocation index corresponding to the above target object according to the product of the first reference index corresponding to the above target object and the preference parameter corresponding to the above target object;

[0128] Determine a second allocation index corresponding to the above target object according to the product of the second reference index corresponding to the above target object and the preference parameter corresponding to the above target object.

[0129] In one embodiment, the above virtual resource allocation module 702 is used to perform the following operations:

[0130] Extract target feature data from the basic feature data corresponding to the above target object, where the target feature data includes at least one of the following features: the attribute feature corresponding to the above target object, the interaction feature generated by the above target object performing an operation on the above target media object;

[0131] Perform logistic regression processing on the above target feature data to obtain an object preference parameter corresponding to the above target object.

[0132] In one embodiment, the above virtual resource allocation module 702 is used to perform the following operations:

[0133] Obtain the sample basic feature data corresponding to the sample object and the preference label of the sample object for the sample media object;

[0134] Extract sample feature data from the above sample basic feature data, where the sample feature data includes at least one of the following features: the attribute feature corresponding to the above sample object, the interaction feature generated by the above sample object performing an operation on the above sample media object;

[0135] Input the above sample feature data into the logistic regression model, trigger the logistic regression model to calculate a preference mapping value according to the above sample feature data and a preset mapping formula, and perform normalization processing on the above preference mapping value to obtain a preference prediction value, where the above preset mapping formula is an exponential function including trainable parameters;

[0136] Adjust the value of the above trainable parameters according to the difference between the above preference prediction value and the above preference label until the training stop condition is reached.

[0137] In one embodiment, the above virtual resource allocation module 702 is configured to perform the following operations:

[0138] When presenting the above target media object to the above virtual resource allocation object, send virtual resources to the above virtual resource allocation object.

[0139] It should be noted that, when the device provided in the above embodiment realizes its functions, only the above division of each functional module is used for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device provided in the above embodiment and the method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be elaborated here.

[0140] Please refer to Figure 8 It shows a structural block diagram of a computer device provided in an embodiment of the present application. The computer device can be a server for executing the above virtual resource allocation method. Specifically:

[0141] The computer device 1100 includes a central processing unit (CPU) 1101, a system memory 1104 including a random access memory (RAM) 1102 and a read only memory (ROM) 1103, and a system bus 1105 connecting the system memory 1104 and the central processing unit 1101. The computer device 1100 also includes a basic input / output system (I / O (Input / Output) system) 1106 for helping to transmit information between various components in the computer, and a mass storage device 1107 for storing an operating system 1113, application programs 1114 and other program modules 1115.

[0142] The basic input / output system 1106 includes a display 1108 for displaying information and input devices 1109 such as a mouse, keyboard, etc. for inputting information about objects. Both the display 1108 and the input devices 1109 are connected to the central processing unit 1101 through an input / output controller 1110 connected to the system bus 1105. The basic input / output system 1106 may also include an input / output controller 1110 for receiving and processing inputs from a plurality of other devices such as a keyboard, mouse, or electronic stylus. Similarly, the input / output controller 1110 also provides outputs to a display screen, printer, or other types of output devices.

[0143] The mass storage device 1107 is connected to the central processing unit 1101 through a mass storage controller (not shown) connected to the system bus 1105. The mass storage device 1107 and its associated computer-readable medium provide non-volatile storage for the computer device 1100. That is to say, the mass storage device 1107 may include computer-readable media (not shown) such as a hard disk or a CD-ROM (Compact Disc Read-Only Memory) drive.

[0144] Without loss of generality, computer-readable media may include computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented by any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes RAM, ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash memory or other solid-state memory technologies, CD-ROM, DVD (Digital Video Disc) or other optical storage, magnetic tape cartridges, tapes, disk storage or other magnetic storage devices. Of course, those skilled in the art will know that computer storage media is not limited to the above several. The above-mentioned system memory 1104 and mass storage device 1107 may be collectively referred to as memory.

[0145] In accordance with various embodiments of the present application, the computer device 1100 may also operate by connecting to a remote computer on a network such as the Internet. That is, the computer device 1100 may be connected to the network 1112 through the network interface unit 1111 connected to the system bus 1105, or in other words, the network interface unit 1111 may also be used to connect to other types of networks or remote computer systems (not shown).

[0146] The above-mentioned memory further includes a computer program, which is stored in the memory and configured to be executed by one or more processors to implement the above-mentioned virtual resource allocation method.

[0147] In an exemplary embodiment, a computer-readable storage medium is further provided. At least one instruction, at least one segment of program, a code set or an instruction set is stored in the above-mentioned storage medium. When the at least one instruction, the at least one segment of program, the code set or the instruction set is executed by a processor, the above-mentioned virtual resource allocation method is implemented.

[0148] Specifically, the virtual resource allocation method includes:

[0149] Obtain the basic feature data corresponding to the target object;

[0150] According to the basic feature data corresponding to the above-mentioned target object, determine the object preference parameter corresponding to the above-mentioned target object, and the object preference parameter indicates the preference degree of the above-mentioned target object for the target media object;

[0151] According to the object preference parameter corresponding to the above-mentioned target object, respectively determine the first allocation index and the second allocation index corresponding to the above-mentioned target object. The first allocation index indicates the probability that the above-mentioned target object performs a preset interaction operation on the above-mentioned target media object in the case of allocating virtual resources, and the second allocation index indicates the probability that the above-mentioned target object performs a preset interaction operation on the above-mentioned target media object in the case of not allocating virtual resources; According to the difference between the first allocation index and the second allocation index, determine the virtual resource allocation indication parameter of the above-mentioned target object for the above-mentioned target media object;

[0152] Perform virtual resource allocation according to the virtual resource allocation indication parameter of the above-mentioned target object for the above-mentioned target media object.

[0153] In one embodiment, the above-mentioned target object is any object in the object set. The above-mentioned performing virtual resource allocation according to the virtual resource allocation indication parameter of the above-mentioned target object for the above-mentioned target media object includes:

[0154] Determine the virtual resource allocation indication parameters of other objects in the above-mentioned object set for the above-mentioned target media object;

[0155] Determine the objects in the above object set whose virtual resource allocation parameters meet the first condition as virtual resource allocation objects;

[0156] Allocate virtual resources to the above virtual resource allocation objects.

[0157] In one embodiment, the above target media object is any media object in the media object set, and the virtual resource allocation based on the virtual resource allocation indication parameter of the above target object for the above target media object includes:

[0158] Determine the virtual resource allocation indication parameters of the above target object for other media objects in the media object set;

[0159] Determine the media objects in the above media object set whose corresponding virtual resource allocation parameters meet the second condition as associated media objects. The above associated media objects are media objects associated with virtual resources, and the above virtual resources are display resources;

[0160] Display the above associated media objects to the above target object.

[0161] In one embodiment, the above determining the first allocation index and the second allocation index corresponding to the above target object according to the object preference parameter corresponding to the above target object includes:

[0162] Obtain the first reference index corresponding to the above target object. The above first reference index indicates the probability that the above target object performs a preset interaction behavior in the case of virtual resource allocation;

[0163] Obtain the second reference index corresponding to the above target object. The above second reference index indicates the probability that the above target object performs a preset interaction behavior in the case of no virtual resource allocation;

[0164] Determine the first allocation index corresponding to the above target object according to the first reference index corresponding to the above target object and the preference parameter corresponding to the above target object;

[0165] Determine the second allocation index corresponding to the above target object according to the second reference index corresponding to the above target object and the preference parameter corresponding to the above target object.

[0166] In one embodiment, the above determining the first allocation index corresponding to the above target object according to the first reference index corresponding to the above target object and the preference parameter corresponding to the above target object includes: determining the first allocation index corresponding to the above target object according to the product of the first reference index corresponding to the above target object and the preference parameter corresponding to the above target object;

[0167] Determining the second allocation index corresponding to the target object according to the second reference index corresponding to the target object and the preference parameter corresponding to the target object includes: determining the second allocation index corresponding to the target object according to the product of the second reference index corresponding to the target object and the preference parameter corresponding to the target object.

[0168] In one embodiment, determining the object preference parameter corresponding to the target object according to the basic feature data corresponding to the target object includes:

[0169] Extracting target feature data from the basic feature data corresponding to the target object, where the target feature data includes at least one of the following types of features: attribute features corresponding to the target object, interaction features generated by the target object performing an operation on the target media object;

[0170] Performing logistic regression processing on the target feature data to obtain the object preference parameter corresponding to the target object.

[0171] In one embodiment, the logistic regression processing is performed by a logistic regression model, and the logistic regression model is trained by the following method:

[0172] Obtaining the sample basic feature data corresponding to the sample object and the preference label of the sample object for the sample media object;

[0173] Extracting sample feature data from the sample basic feature data, where the sample feature data includes at least one of the following types of features: attribute features corresponding to the sample object, interaction features generated by the sample object performing an operation on the sample media object;

[0174] Inputting the sample feature data into the logistic regression model, triggering the logistic regression model to calculate a preference mapping value according to the sample feature data and a preset mapping formula, and performing normalization processing on the preference mapping value to obtain a preference prediction value, where the preset mapping formula is an exponential function including trainable parameters;

[0175] Adjusting the value of the trainable parameter according to the difference between the preference prediction value and the preference label until the training stop condition is reached.

[0176] In one embodiment, allocating virtual resources to the virtual resource allocation object includes: when presenting the target media object to the virtual resource allocation object, sending virtual resources to the virtual resource allocation object.

[0177] Optionally, the computer-readable storage medium may include: ROM (Read Only Memory), RAM (Random Access Memory), SSD (Solid State Drives), or optical discs, etc. Among them, the random access memory may include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).

[0178] In an exemplary embodiment, a computer program product or a computer program is further provided. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the above virtual resource allocation method.

[0179] It should be understood that the "plurality" mentioned herein refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after. In addition, the step numbers described herein only exemplarily show a possible execution sequence between steps. In some other embodiments, the above steps may not be executed in the order of the numbers. For example, two steps with different numbers are executed simultaneously, or two steps with different numbers are executed in the reverse order of the illustration. The embodiments of the present application do not limit this.

[0180] In the embodiments of the present application, the term "module" or "unit" refers to a computer program with a predetermined function or a part of a computer program, which works with other related parts to achieve a predetermined goal, and can be fully or partially implemented by using software, hardware (such as a processing circuit or a memory), or a combination thereof. Similarly, one processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of the overall module or unit that includes the function of the module or unit.

[0181] In addition, in the specific implementation manners of the present application, for data related to object information, etc., when the above embodiments of the present application are applied to specific products or technologies, object permission or consent needs to be obtained, and the collection, use, and processing of the relevant data need to comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0182] The above are only exemplary embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A virtual resource allocation method, characterized in that, The method includes: Obtaining basic feature data corresponding to a target object; Determining an object preference parameter corresponding to the target object according to the basic feature data corresponding to the target object, where the object preference parameter indicates the preference degree of the target object for a target media object; Respectively determining a first allocation index and a second allocation index corresponding to the target object according to the object preference parameter corresponding to the target object, where the first allocation index indicates the probability that the target object performs a preset interaction operation on the target media object when virtual resources are allocated, and the second allocation index indicates the probability that the target object performs a preset interaction operation on the target media object when virtual resources are not allocated; determining a virtual resource allocation indication parameter of the target object for the target media object according to the difference between the first allocation index and the second allocation index; Performing virtual resource allocation according to the virtual resource allocation indication parameter of the target object for the target media object.

2. The method according to claim 1, characterized in that The target object is any object in an object set, and performing virtual resource allocation according to the virtual resource allocation indication parameter of the target object for the target media object includes: Determining virtual resource allocation indication parameters of other objects in the object set for the target media object; Determining objects in the object set whose virtual resource allocation parameters meet a first condition as virtual resource allocation objects; Allocating virtual resources to the virtual resource allocation objects.

3. The method according to claim 1, characterized in that, The target media object is any media object in a media object set, and performing virtual resource allocation according to the virtual resource allocation indication parameter of the target object for the target media object includes: Determining virtual resource allocation indication parameters of the target object for other media objects in the media object set; Determining media objects in the media object set whose corresponding virtual resource allocation parameters meet a second condition as associated media objects, where the associated media objects are media objects associated with virtual resources, and the virtual resources are display resources; Displaying the associated media objects to the target object.

4. The method according to any one of claims 1 to 3, characterized in that Respectively determining the first allocation index and the second allocation index corresponding to the target object according to the object preference parameter corresponding to the target object includes: Obtaining a first reference index corresponding to the target object, where the first reference index indicates the probability that the target object performs a preset interaction behavior when virtual resources are allocated; Obtaining a second reference index corresponding to the target object, where the second reference index indicates the probability that the target object performs a preset interaction behavior when virtual resources are not allocated; Determining the first allocation index corresponding to the target object according to the first reference index corresponding to the target object and the preference parameter corresponding to the target object; Determining the second allocation index corresponding to the target object according to the second reference index corresponding to the target object and the preference parameter corresponding to the target object.

5. The method according to claim 4, wherein Determining the first allocation index corresponding to the target object according to the first reference index corresponding to the target object and the preference parameter corresponding to the target object includes: determining the first allocation index corresponding to the target object according to the product of the first reference index corresponding to the target object and the preference parameter corresponding to the target object; Determining the second allocation index corresponding to the target object according to the second reference index corresponding to the target object and the preference parameter corresponding to the target object includes: determining the second allocation index corresponding to the target object according to the product of the second reference index corresponding to the target object and the preference parameter corresponding to the target object.

6. The method according to any one of claims 1 to 3, characterized in that, Determining the object preference parameter corresponding to the target object according to the basic feature data corresponding to the target object includes: Extracting target feature data from the basic feature data corresponding to the target object, where the target feature data includes at least one of the following types of features: the attribute feature corresponding to the target object, the interaction feature generated by the target object performing an operation on the target media object; Performing logistic regression processing on the target feature data to obtain the object preference parameter corresponding to the target object.

7. The method according to claim 6, characterized in that, The logistic regression processing is performed by a logistic regression model, and the logistic regression model is trained by the following method: Obtaining the sample basic feature data corresponding to the sample object and the preference label of the sample object for the sample media object; Extracting sample feature data from the sample basic feature data, where the sample feature data includes at least one of the following types of features: the attribute feature corresponding to the sample object, the interaction feature generated by the sample object performing an operation on the sample media object; Inputting the sample feature data into the logistic regression model, triggering the logistic regression model to calculate a preference mapping value according to the sample feature data and a preset mapping formula, and performing normalization processing on the preference mapping value to obtain a preference prediction value, where the preset mapping formula is an exponential function including trainable parameters; Adjusting the value of the trainable parameter according to the difference between the preference prediction value and the preference label until the training stop condition is reached.

8. The method according to claim 2, wherein Allocating virtual resources to the virtual resource allocation object includes: when presenting the target media object to the virtual resource allocation object, sending virtual resources to the virtual resource allocation object.

9. A virtual resource allocation device, characterized in that, The device includes: A data acquisition module for acquiring the basic feature data corresponding to the target object; A virtual resource allocation module for performing the following operations: Determine the object preference parameter corresponding to the target object according to the basic feature data corresponding to the target object, where the object preference parameter indicates the preference degree of the target object for the target media object; determine the first allocation index and the second allocation index corresponding to the target object respectively according to the object preference parameter corresponding to the target object, where the first allocation index indicates the probability that the target object performs a preset interaction operation on the target media object in the case of allocating virtual resources, and the second allocation index indicates the probability that the target object performs a preset interaction operation on the target media object in the case of not allocating virtual resources; determine the virtual resource allocation indication parameter of the target object for the target media object according to the difference between the first allocation index and the second allocation index; and perform virtual resource allocation according to the virtual resource allocation indication parameter of the target object for the target media object.

10. A computer device, characterized in that, The computer device includes a processor and a memory, and at least one instruction, at least one program, a code set or an instruction set is stored in the memory, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the virtual resource allocation method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, At least one instruction, at least one program, a code set or an instruction set is stored in the storage medium, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the virtual resource allocation method according to any one of claims 1 to 8.

Images