Multimedia Resource Control Method, Device, Electronic Device, and Storage Medium

Abstract

The present application provides a method, apparatus, electronic device and storage medium for regulating multimedia resources. The method includes: determining an estimated click-through rate range of the multimedia resource; obtaining an estimated click-through rate range - selection probability comparison table, where the estimated click-through rate range - selection probability comparison table includes the corresponding relationships between multiple estimated click-through rate ranges and multiple selection probabilities; determining the selection probability corresponding to the estimated click-through rate range according to the estimated click-through rate range and the estimated click-through rate range - selection probability comparison table; obtaining a first random factor; and regulating the exposure of the multimedia resource according to the first random factor and the selection probability. In the embodiments of the present application, different estimated click-through rate ranges have corresponding selection probabilities, so that multimedia resources with low or high estimated click-through rates have the opportunity to be exposed; dynamically adjusting the selection probability corresponding to the estimated click-through rate range can achieve uniform exposure of the multimedia resources and improve the conversion rate of the multimedia resources.

Description

Technical Field

[0001] This application relates to the fields of computer technology and information processing technology, and particularly to a method, device, electronic device, and storage medium for regulating and controlling multimedia resources. Background Art

[0002] With the continuous development of the Internet, multimedia resources have become increasingly complex and numerous. When a service platform receives a service request, it needs to recommend multimedia resources that meet the service request to the user. Based on the task requirements of the service platform, certain multimedia resources need to meet a certain exposure volume within a preset time period.

[0003] However, the prior art is difficult to dynamically adjust according to the actual exposure situation and often requires manual control to complete the target task. Summary of the Invention

[0004] Embodiments of this application provide a method, device, electronic device, and storage medium for regulating and controlling multimedia resources to solve the problems existing in the related art. The technical solutions are as follows:

[0005] In a first aspect, embodiments of this application provide a method for regulating and controlling multimedia resources, including:

[0006] Determine the estimated click-through rate interval of the multimedia resource;

[0007] Obtain an estimated click-through rate interval - selection probability comparison table for the multimedia resource; wherein, the estimated click-through rate interval - selection probability comparison table includes the corresponding relationships between multiple estimated click-through rate intervals and multiple selection probabilities;

[0008] According to the estimated click-through rate interval and the estimated click-through rate interval - selection probability comparison table, determine the selection probability corresponding to the estimated click-through rate interval;

[0009] Obtain a first random factor;

[0010] Regulate the exposure of the multimedia resource according to the first random factor and the selection probability.

[0011] In a second aspect, embodiments of this application provide a device for regulating and controlling multimedia resources, including:

[0012] An estimated click-through rate interval determination module, configured to determine the estimated click-through rate interval of the multimedia resource;

[0013] A comparison table acquisition module, configured to obtain an estimated click-through rate interval - selection probability comparison table for the multimedia resource; wherein, the estimated click-through rate interval - selection probability comparison table includes the corresponding relationships between multiple estimated click-through rate intervals and multiple selection probabilities;

[0014] A selection probability determination module, configured to determine a selection probability corresponding to the estimated click-through rate range according to an estimated click-through rate range - selection probability look-up table;

[0015] A first random factor acquisition module, configured to acquire a first random factor;

[0016] A regulation module, which regulates the exposure of the multimedia resource according to the first random factor and the selection probability.

[0017] Thirdly, an embodiment of the present application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, so that the at least one processor can execute the above-mentioned multimedia resource regulation method.

[0018] Fourthly, an embodiment of the present application provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are run on a computer, the method in any one of the above aspects is executed.

[0019] The advantages or beneficial effects in the above technical solutions at least include:

[0020] In the embodiment of the present application, when a service request for exposing a multimedia resource can be received, the estimated click-through rate range of the multimedia resource is acquired, and based on the obtained estimated click-through rate range - selection probability look-up table, the selection probability for the multimedia resource to be selected for exposure is determined; then, the exposure of the multimedia resource is regulated through the first random factor and the selection probability.

[0021] The multimedia resource regulation method provided by the embodiment of the present application determines the selection probability for the multimedia resource to be selected for exposure through an estimated click-through rate range - selection probability look-up table. Each estimated click-through rate range has a corresponding selection probability, so that multimedia resources with low or high estimated click-through rates all have the opportunity to be exposed. Moreover, the selection probability corresponding to the estimated click-through rate range can be dynamically adjusted according to the exposure situation, thereby dynamically adjusting the exposure of the multimedia resource, achieving uniform exposure of the multimedia resource, and improving the conversion rate of the multimedia resource.

[0022] The above summary is only for the purpose of the specification and is not intended to be limiting in any way. In addition to the above-described illustrative aspects, embodiments and features, further aspects, embodiments and features of the present application will be readily apparent by reference to the drawings and the following detailed description. Description of the Drawings

[0023] In the accompanying drawings, unless otherwise specified, the same reference numerals in multiple drawings denote the same or similar components or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments disclosed in the present application and should not be regarded as limiting the scope of the present application.

[0024] Figure 1 Schematically shows a schematic diagram of an environmental application according to an embodiment of the present application.

[0025] Figure 2 Is a flowchart showing the method for regulating multimedia resources according to an embodiment of the present application.

[0026] Figure 3 Is Figure 2 A sub - flowchart diagram of step S210 in

[0027] Figure 4 Is Figure 3 A sub - flowchart diagram of step S212 in

[0028] Figure 5 Is Figure 2 A sub - flowchart diagram of step S220 in

[0029] Figure 6 Is Figure 5 A sub - flowchart diagram of step S224 in

[0030] Figure 7 Is Figure 2 A sub - flowchart diagram of step S250 in

[0031] Figure 8 Is a schematic diagram of a multimedia resource regulation device according to an embodiment of the present application.

[0032] Figure 9 Is a block diagram of an electronic device for implementing the multimedia resource regulation method of the embodiment of the present application. Detailed implementation manners

[0033] In the following, only some exemplary embodiments are briefly described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present application. Therefore, the drawings and the description are considered to be exemplary in nature rather than restrictive.

[0034] It should be noted that in the embodiments of the present application, the descriptions involving "first", "second", etc. are only for descriptive purposes, and should not be construed as indicating or implying their relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. Additionally, the technical solutions between various embodiments may be combined with each other, but it must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present application.

[0035] For the regulation and control of multimedia resources, traditional methods include the probability random method and the method of setting a predicted click-through rate threshold. The probability random method estimates the total exposure volume of multimedia resources in the inventory, calculates the ratio of the exposure volume of each multimedia resource to the total exposure volume respectively, and then uses this ratio as a measurement criterion to regulate the exposure of multimedia resources. The method of setting a predicted click-through rate threshold sets a threshold range of the predicted click-through rate according to empirical data, and regulates the exposure of the multimedia resource when the predicted click-through rate meets this threshold range.

[0036] Traditional probability random method and the method of setting a predicted click-through rate threshold have the following defects:

[0037] First, the business traffic is unstable and cannot be dynamically adjusted according to the actual exposure situation to complete the target volume. To complete the target volume, it usually relies on manual regulation.

[0038] Second, it is difficult to achieve uniform exposure of multimedia resources. Multimedia resources may quickly complete the target volume within a relatively short time period, and it is difficult to obtain good exposure effects, such as low click-through rate or low conversion rate.

[0039] Third, it cannot meet the targeted exposure of multimedia resources with targeted requirements.

[0040] Fourth, it cannot obtain the expected click-through rate level through regulation.

[0041] The present application provides multiple embodiments to solve the above defects, as specifically described below.

[0042] In the description of the present application, it should be understood that the numerical labels before the steps do not identify the sequence of execution of the steps, but are only used to facilitate the description of the present application and to distinguish each step, and thus should not be construed as a limitation to the present application.

[0043] The following are the term explanations of the present application:

[0044] Predicted Click-Through Rate (PCTR): To achieve better exposure effects for multimedia resources, the click-through rate of a multimedia resource is predicted before its exposure. If the prediction is relatively accurate, when multiple multimedia resources under the same request compete, selecting the multimedia resource with a higher predicted click-through rate for exposure will bring better benefits. The predicted click-through rate can be predicted based on various factors, such as historical click-through rates, requested sites, requesting users, etc.

[0045] Click-Through Rate (CTR): The ratio of the number of clicks on a multimedia resource to the number of exposures of the multimedia resource, which is an important factor in measuring the exposure effect of a multimedia resource. The higher the click-through rate, that is, the more times the multimedia resource is clicked, the better the exposure effect.

[0046] Figure 1 Schematically shows an environmental application schematic diagram according to an embodiment of the present application. As Figure 1 shown:

[0047] The provider network 2 can connect multiple mobile terminals 6 through the network 4. The provider network 2 can achieve the exposure of multimedia resources. The provider network 2 can also provide specific content services and can achieve the exposure of multimedia resources when providing specific content services. For example, when a user searches for content, according to the search information, multimedia resources related to the search information are exposed; for example, when a user browses web content or browses a game page, multimedia resources are exposed in the form of pop-up ads; for another example, when a user watches a video, multimedia resources are inserted between frames of the video to achieve the exposure of multimedia resources.

[0048] The provider network 2 can be located in a data center such as a single location or distributed in different geographical locations (for example, in multiple locations). The provider network 2 can provide services via one or more networks 4. The network 4 includes various network devices, such as routers, switches, multiplexers, hubs, modems, bridges, repeaters, firewalls, proxy devices, and / or the like. The network 4 can include physical links, such as coaxial cable links, twisted pair cable links, fiber optic links, combinations thereof, etc. The network 4 can include wireless links, such as cellular links, satellite links, Wi-Fi links, etc.

[0049] The provider network 2 can be configured to receive multiple multimedia resources, as well as receive the target volume, targeting information, or target number of clicks of the multimedia resources.

[0050] The provider network 2 can be configured to regulate the exposure of multimedia resources; so that the number of exposures of the multimedia resources meets the approximate quantity requirement, the targeted exposure requirement, or the target click-through number requirement. And when the multimedia resources meet the approximate quantity requirement, they are evenly exposed to improve the exposure effect. The provider network 2 manages various data information for regulating multimedia resources and performs regulation based on the various data information.

[0051] The provider network 2 can be implemented by one or more computing nodes. The one or more computing nodes can include virtualized computing instances. The virtualized computing instances can include virtual machines, such as emulations of computer systems, operating systems, servers, etc. The computing nodes can load virtual machines based on virtual images and / or other data defining specific software (e.g., operating system, dedicated application, server) for emulation. As the demand for different types of processing services changes, different virtual machines can be loaded and / or terminated on the one or more computing nodes. A hypervisor can be implemented to manage the use of different virtual machines on the same computing node.

[0052] Multiple mobile terminals 6 can be configured to access the content services and multimedia resources of the provider network 2. The multiple mobile terminals 6 can include any type of electronic device, such as mobile devices, tablet devices, laptop computers, workstations, virtual reality devices, gaming devices, set-top boxes, digital streaming devices, vehicle terminals, smart TVs, set-top boxes, etc.

[0053] The multiple mobile terminals 6 can also be configured to send multimedia resource content, as well as the approximate quantity of exposure, the target click-through number, or the targeted information of the multimedia resources. The provider network 2 can receive the multimedia resource content sent by the multiple mobile terminals 6, as well as the approximate quantity of exposure, the target click-through number, or the targeted exposure information of the multimedia resources, and regulate the exposure of the multimedia resources so that the exposure of the multimedia resources meets the approximate quantity of exposure, the target click-through number, or the targeted exposure, etc.

[0054] The following will introduce the solution for the provider network 2 to regulate the exposure of multimedia resources through multiple embodiments. This solution can be implemented by a computer device 1300, which can be the provider network 2 or its computing node, or a mobile terminal 6.

[0055] Figure 2 A flowchart showing a multimedia resource regulation method according to an embodiment of the present application is as follows Figure 2 As shown, the multimedia resource regulation method can include:

[0056] S210, determining the estimated click-through rate interval of the multimedia resource.

[0057] In the embodiments of the present application, the predicted click-through rate of the multimedia resource may be any value between 0 and 1 or between 0 and 100%, that is, the range of the predicted click-through rate may be 0 to 1 or 0 to 100%. The range of the predicted click-through rate is divided into intervals including multiple smaller predicted click-through rate ranges. The number of predicted click-through rate intervals may be 30, 50, 80, or 100.

[0058] In one example, the range of the predicted click-through rate may also be a preset range, such as 10% to 90%. By dividing this preset range into multiple predicted click-through rate intervals, data processing with low probability can be avoided, and the efficiency of data processing can be improved.

[0059] In the embodiments of the present application, the number of predicted click-through rate intervals is not limited. Of course, the number of predicted click-through rate intervals may be a preset value, so as to divide the predicted click-through rate into a preset number of predicted click-through rate intervals according to the preset value.

[0060] In one example, when the range of the predicted click-through rate is 0 to 1, the predicted click-through rate can be divided into 50 intervals, for example: 0 to 0.0014; 0.0014 to 0.0023; 0.0023 to 0.0048; 0.0048 to 0.0105... 0.092 to 0.10; 0.10 to 0.16; 0.16 to 0.28; 0.28 to 0.42; 0.42 to 1.

[0061] It can be understood that the multiple predicted click-through rate intervals may be unevenly divided. The multiple predicted click-through rate intervals may be divided according to the principle that the exposure quantities of multiple intervals in the first historical data are equal; or they may be divided according to the principle that the click numbers of multiple intervals are equal.

[0062] For different multimedia resources, there are different ways to divide the predicted click-through rate intervals. When the business platform receives a multimedia resource, it will determine multiple predicted click-through rate intervals of the multimedia resource according to the first historical data of other multimedia resources similar to the multimedia resource. It may also update multiple predicted click-through rate intervals according to the first historical data of the multimedia resource itself.

[0063] To determine the predicted click-through rate interval of a multimedia resource, it may be to first determine the predicted click-through rate of the multimedia resource, and then determine the predicted click-through rate interval into which the predicted click-through rate falls.

[0064] S220. Obtain a predicted click-through rate interval - selection probability comparison table. Among them, the predicted click-through rate interval - selection probability comparison table includes the corresponding relationship between multiple predicted click-through rate intervals and multiple selection probabilities.

[0065] The selection probability can be understood as the probability of selecting the exposure of this multimedia resource. The greater the selection probability, the greater the possibility of exposure.

[0066] In the service platform, there are a large number of multimedia resources that need to be exposed, and each multimedia resource has corresponding exposure quantity requirements. By regulating the exposure of multimedia resources through the selection probability, the required exposure quantity of multiple multimedia resources can be completed.

[0067] It can be understood that the selection probability is related to the exposure quantity requirement, that is, the target exposure volume. The more the target exposure volume, the greater the selection probability, so as to complete the target exposure volume.

[0068] In the embodiments of this application, after dividing the estimated click-through rate into multiple estimated click-through rate intervals, corresponding selection probabilities are respectively configured for each estimated click-through rate interval, and the selection probability is refined to different estimated click-through rate intervals. In the interval with a higher estimated click-through rate, when the selection probability is large, the click-through rate obtained after the exposure of the multimedia resource is greater.

[0069] In the embodiments of this application, based on the estimated click-through rate interval - selection probability look-up table, the exposure of multimedia resources is regulated. Not only can the exposure volume of multimedia resources be regulated to complete the target exposure volume, but also the click-through rate of multimedia resources can be regulated to make the click-through rate of multimedia resources controllable.

[0070] S230. According to the estimated click-through rate interval and the estimated click-through rate interval - selection probability look-up table, determine the selection probability corresponding to the estimated click-through rate interval.

[0071] After determining the estimated click-through rate interval of the multimedia resource, then look up the estimated click-through rate interval - selection probability look-up table, and the selection probability corresponding to this estimated click-through rate interval can be obtained.

[0072] S240. Obtain a first random factor. The first random factor is a random number generated according to a preset rule.

[0073] The multimedia resource regulation method in the embodiments of this application can be executed in response to the service request when the service request is received.

[0074] When the service request is received, a first random factor can be obtained. The first random factor is a randomly generated random number. The random number can be a percentage (or a decimal greater than 0 and less than 1); it can also be other forms of data, such as a positive integer, and then converted into a percentage (or a decimal greater than 0 and less than 1) according to a preset conversion rule.

[0075] After obtaining the first random factor, the first random factor can be associated with the service request so as to obtain the first random factor synchronously in response to multiple service requests.

[0076] S250, adjust the exposure of the multimedia resource according to the first random factor and the selection probability.

[0077] In one example, when the first random factor is greater than or equal to the selection probability, it can be determined that the multimedia resource can be exposed, that is, the multimedia resource has the opportunity to be exposed.

[0078] In the embodiments of the present application, the exposure of the multimedia resource may refer to sending and displaying the multimedia resource on the display interface of the client, so that the user using the client can see the multimedia resource and realize the dissemination of the multimedia resource.

[0079] In one example, when the first random factor is less than or equal to the selection probability, it can be determined that the multimedia resource can be exposed, that is, the multimedia resource has the opportunity to be exposed.

[0080] Determine multiple multimedia resources with the opportunity to be exposed through the first random factor and the selection probability, and then determine the finally exposed multimedia resources from the multiple multimedia resources according to the target quantity, so as to realize the regulation of the multimedia resources.

[0081] In the embodiments of the present application, when receiving a service request for exposing multimedia resources, obtain the estimated click-through rate interval of the multimedia resource, and determine the selection probability of the multimedia resource being selected for exposure based on the obtained estimated click-through rate interval - selection probability comparison table; then adjust the exposure of the multimedia resource through the first random factor and the selection probability.

[0082] The multimedia resource regulation method provided by the embodiments of the present application determines the selection probability of the multimedia resource being selected for exposure through the estimated click-through rate interval - selection probability comparison table. Different estimated click-through rate intervals have corresponding selection probabilities, so that multimedia resources with low or high estimated click-through rates all have the opportunity to be exposed. Moreover, the selection probability corresponding to the estimated click-through rate interval can be dynamically adjusted according to the exposure situation, so as to dynamically adjust the exposure of the multimedia resource, realize the uniform exposure of the multimedia resource, and improve the conversion rate of the multimedia resource.

[0083] In one implementation, as Figure 3 shown, step S210 includes:

[0084] S211, obtain the interval set of the multimedia resource, and the interval set includes multiple estimated click-through rate intervals divided according to a preset rule.

[0085] The interval set of the multimedia resource may refer to the set of multiple estimated click-through rate intervals of the multimedia resource; the estimated click-through rate of the multimedia resource will fall into one of the estimated click-through rate intervals in the interval set.

[0086] The interval set of multimedia resources can be fixed or dynamically updated.

[0087] In one example, when obtaining a target task, the interval set of multimedia resources is updated until the target task is completed, and the interval set of the multimedia resources remains unchanged.

[0088] S212. Obtain the target estimated click-through rate of the multimedia resource.

[0089] The target estimated click-through rate of the multimedia resource represents the proportion of the number of clicks that are estimated to occur after the multimedia resource is exposed.

[0090] The target estimated click-through rate of the multimedia resource can be related to the multimedia resource itself. For example, it is related to the text popularity, picture quality, picture theme, etc. of the multimedia resource. The higher the text popularity, the higher the target estimated click-through rate; the better the picture quality, the higher the target estimated click-through rate; the more attractive the picture theme, the higher the target estimated click-through rate.

[0091] The target estimated click-through rate of the multimedia resource can also be related to the service request information. The greater the degree of relevance between the multimedia resource and the service request, the higher the target estimated click-through rate. Therefore, for different service requests, the target estimated click-through rate of the same multimedia resource may be different.

[0092] S213. Determine that the estimated click-through rate interval in which the target estimated click-through rate falls is the estimated click-through rate interval of the multimedia resource.

[0093] In one example, the multiple estimated click-through rate intervals of the multimedia resource include 0 - 0.0014; 0.0014 - 0.0023; 0.0023 - 0.0048; 0.0048 - 0.0105... 0.092 - 0.10; 0.10 - 0.16; 0.16 - 0.28; 0.28 - 0.42; 0.42 - 1. When the target estimated click-through rate of the multimedia resource is 0.12, determine that the estimated click-through rate interval of the multimedia resource is 0.10 - 0.16.

[0094] In the embodiments of the present application, by responding to a service request, the target estimated click-through rate of the multimedia resource is obtained, and then the estimated click-through rate interval in which the target estimated click-through rate falls is determined, so as to control the exposure situation of the multimedia resource according to the estimated click-through rate interval of the multimedia resource.

[0095] In one implementation manner, step S211 includes: obtaining first historical data, where the first historical data includes historical estimated click-through rates and historical exposure amounts corresponding to the historical estimated click-through rates; based on the historical estimated click-through rates and historical exposure amounts, determining multiple estimated click-through rate intervals of the multimedia resource, where the historical exposure amounts of the multiple estimated click-through rate intervals are equal.

[0096] The first historical data may be the historical data of other multimedia resources similar to the multimedia resource; it may also be the historical data of the multimedia resource itself. The multimedia resource and the multimedia resource of the first historical data may have the same category, the same field, or the same first estimated click-through rate, etc. The first estimated click-through rate is the estimated click-through rate determined according to the content information of the multimedia resource.

[0097] In one example, in the first historical data, the total number of exposures within a day is 10,000, including the number of exposures corresponding to each historical estimated click-through rate. Assuming that the interval set of the multimedia resource is determined to be 50 estimated click-through rate intervals, the number of exposures in each estimated click-through rate interval is 200. Then, in the first historical data, within the historical estimated click-through rate range of 0 to 0.0012, 200 have been exposed, so it is determined that 0 to 0.0012 is the first estimated click-through rate interval, and so on, and 50 estimated click-through rate intervals are respectively determined for multimedia resource regulation.

[0098] In the embodiments of the present application, the historical exposure amounts of multiple estimated click-through rate intervals are equal. According to specific actual situations, the historical exposure amounts of multiple estimated click-through rate intervals may also be close, that is, the difference in historical exposure amounts is within a preset allowable range.

[0099] In other embodiments, the estimated click-through rate may also be partitioned according to the principle that the historical click numbers of multiple estimated click-through rate intervals are equal.

[0100] In the embodiments of the present application, the estimated click-through rate is divided into multiple estimated click-through rate intervals through the first historical data. The number of exposures in each estimated click-through rate interval is equal or close, so that the exposure situation in each estimated click-through rate interval is close. It can be made that the range of the estimated click-through rate interval with more exposure opportunities is smaller, and the range of the estimated click-through rate interval with fewer exposure opportunities is larger. In the part where the estimated click-through rate interval is smaller, the estimated click-through rate can be further subdivided to facilitate the regulation of the click-through rate of the multimedia resource.

[0101] In one embodiment, as Figure 4 shown, step S212 includes:

[0102] S2121, determining the first estimated click-through rate of the multimedia resource according to the content information of the multimedia resource.

[0103] The first estimated click-through rate can be static and related to the multimedia resource itself. The content information of the multimedia resource can include the picture content, picture color, text content, or text attractiveness of the multimedia resource, etc. Based on the content information of the multimedia resource, the first estimated click-through rate can be determined. Therefore, when receiving a multimedia resource that needs to be exposed, the first estimated click-through rate of the multimedia resource can be determined and associated with the multimedia resource. This is to facilitate calculating the target estimated click-through rate in combination with the service request when receiving the service request.

[0104] S2122. Determine the second estimated click-through rate of the multimedia resource according to the service request information.

[0105] The service request can be a request sent by the user, such as a search content request. According to the user's search content request, multimedia resources related to the search content request are displayed. The service request information can be the search content keyword. By determining the degree of association between the search content keyword and the multimedia resource keyword, the second estimated click-through rate of the multimedia resource is determined.

[0106] The service request can also be a promotion position preset by the service platform. For example, when the user browses the web content, at preset intervals or under preset conditions, a multimedia resource pops up at a preset position on the user's browsing page to achieve the exposure of the multimedia resource. When the service platform detects that the multimedia resource can pop up, it means that the service request is received. The service request can be browsing content information, etc. The browsing content information can include video content, text content, and specific classification content. The specific classification content can be classified as related to infants and children, health consultation, entertainment, or learning, etc.

[0107] The service request information can also be the relevant service information carried in the service request, such as user information, etc. The user information can include user gender, user age, user location, user click-through rate, etc.

[0108] Determining the second estimated click-through rate of the multimedia resource according to the service request can be to determine the second estimated click-through rate based on the degree of association between the content of the multimedia resource and the service request information. It can be understood that the higher the degree of association between the content of the multimedia resource and the service request information, the greater the second estimated click-through rate. Conversely, the lower the degree of association between the content of the multimedia resource and the service request information, the smaller the second estimated click-through rate.

[0109] In one example, when a multimedia resource is received, multiple information tags can be added to the multimedia resource. When the service request information includes tags related to the information tags, a second estimated click-through rate is calculated. When the service request information does not include tags related to the information tags, the second estimated click-through rate is determined to be 0, or the second estimated click-through rate is determined to be an extremely small value, such as 0.00009, and the specific value is not limited.

[0110] In one example, when a service request is received, the service request can be parsed to obtain tags identifying the service request information; for example, the specific classification content of the service request is parsed and obtained.

[0111] S2123. Obtain a target estimated click-through rate according to the first estimated click-through rate and / or the second estimated click-through rate.

[0112] Obtain a target estimated click-through rate according to the first estimated click-through rate and the second estimated click-through rate; for example, the first estimated click-through rate can be multiplied by the second estimated click-through rate to obtain the target estimated click-through rate.

[0113] The target estimated click-through rate can also be equal to the second estimated click-through rate, that is, the first estimated click-through rate is not considered.

[0114] The target estimated click-through rate can also be equal to the first estimated click-through rate, that is, the second estimated click-through rate is not considered.

[0115] The embodiments of the present application provide multiple ways to obtain the target estimated click-through rate, and one of them can be adopted according to the actual application situation.

[0116] In the embodiments of the present application, by combining the first estimated click-through rate associated with the multimedia resource itself and the second estimated click-through rate associated with the service request, the target estimated click-through rate is determined, so that the target estimated click-through rate comprehensively considers two factors of the resource itself and the service request, and the estimation is more accurate; at the same time, because it is related to the service request, different target estimated click-through rates are obtained for different service requests, which is convenient for realizing the regulation of the multimedia resource.

[0117] In one implementation manner, step S220 includes: when the first preset time interval has passed, generating a comparison table of estimated click-through rate intervals - selection probabilities of the multimedia resource based on the remaining amount of the multimedia resource; the remaining amount includes the remaining exposure duration and the remaining exposure amount.

[0118] In the embodiments of the present application, the amount refers to the agreed exposure amount that needs to be completed, for example, how many exposures are completed within a certain period of time to meet the user's requirement for the exposure of the multimedia resource.

[0119] For example, after an interval of 30 minutes, the remaining approximate quantity is to complete 5,000 exposures within the next 5 hours. Then, based on the remaining approximate quantity of 5,000 exposures to be completed within 5 hours, a predicted click-through rate interval - selection probability comparison table is generated to ensure that 5,000 exposures can be completed within the remaining 5 hours.

[0120] Further, after another 30 minutes, 1,200 exposures are completed within this 30 minutes. Then the remaining approximate quantity is to complete 3,800 exposures within the next 4.5 hours. Based on the 3,800 exposures to be completed within 4.5 hours, a predicted click-through rate interval - selection probability comparison table is generated to ensure that 3,800 exposures can be completed within the remaining 4.5 hours.

[0121] In the embodiment of the present application, by updating the predicted click-through rate interval - selection probability comparison table of the multimedia resource at intervals of the first preset time, the exposure situation of the multimedia resource is dynamically adjusted, and the uniform exposure of the multimedia resource is regulated.

[0122] In one implementation manner, as Figure 5 shown, step S220 includes:

[0123] S221, obtaining the target approximate quantity of the multimedia resource, where the target approximate quantity includes a target duration and a target exposure quantity.

[0124] The target approximate quantity is the agreed exposure quantity reached with the user, and it is required to complete the target exposure quantity within the target duration. In order to obtain a good exposure effect, it is necessary to uniformly complete the target exposure quantity within the target duration.

[0125] The target approximate quantity can be obtained by obtaining the corresponding configuration parameters. The configuration parameters can be configured by the staff or the user.

[0126] S222, dividing the target duration into multiple time periods.

[0127] In one example, the durations of the multiple time periods are equal. The duration of the time period can be 10 minutes, 15 minutes, 30 minutes, or 45 minutes, etc. Different multimedia resources can adopt different time periods or the same time period.

[0128] In one example, the target approximate quantity is 10,000 exposures per day. The target duration is one day, and the target exposure quantity is 10,000 times. The target duration is divided into 96 time periods, and the time period is 15 minutes. The time period can be set according to the specific actual situation.

[0129] In the business platform, the time period can be uniformly configured or can be the time period configured separately for specific multimedia resources.

[0130] S223. Determine the sub-exposure amount corresponding to the time period according to the target exposure amount.

[0131] In one example, the target exposure amount is 10,000 exposures per day. The time period is 15 minutes, and there are 96 time periods in a day. Assuming that the sub-exposure amount for each time period is the same, the sub-exposure amount that needs to be exposed for each time period is 10,000 / 96 = 105 exposures. Thus, after 96 time periods, the target exposure amount can be completed.

[0132] Other methods can also be used to determine the sub-exposure amount corresponding to the time period according to the target exposure amount, and the embodiments of the present application do not limit this.

[0133] S224. Based on the sub-exposure amount corresponding to the time period, determine the estimated click-through rate interval - selection probability comparison table corresponding to the time period.

[0134] The greater the selection probability, the greater the chance of the multimedia resource being exposed, and there will be more exposure amounts. Therefore, according to the sub-exposure amount corresponding to the time period, determining the estimated click-through rate interval - selection probability comparison table corresponding to the time period can ensure that the sub-exposure amount of the current time period can be completed within the current time period. By analogy, according to this method, in each time period, it is ensured that the sub-exposure amount of the current time period is completed, so as to ensure that the target exposure amount can be completed and completed evenly.

[0135] In the embodiments of the present application, by respectively determining the sub-exposure amount of each time period, and then determining the estimated click-through rate interval - selection probability comparison table corresponding to the time period according to the sub-exposure amount, the sub-exposure amount corresponding to the time period can be allocated according to the actual specific situation, so that the target exposure amount can gradually reach the target and achieve uniform exposure.

[0136] In one implementation manner, step S223 includes:

[0137] Obtain historical traffic-time data, where the historical traffic-time data includes multiple time periods and the traffic corresponding to each of the multiple time periods;

[0138] Based on the historical traffic-time data, determine the first progress requirements for multiple time periods; the first progress requirement is the progress that needs to be completed for exposure in the corresponding time period;

[0139] Based on the target exposure amount and the first progress requirements of the time period, determine the sub-exposure amounts corresponding to multiple time periods.

[0140] The historical traffic-time data can be the traffic data for a specific time period. The durations of multiple time periods can be the same or different. The traffic data can be expressed as a percentage of the total daily traffic.

[0141] In one example, the durations of multiple time periods are the same, which can be 1 hour. The historical traffic-time data can be:

[0142] Hour 0～1 1～2 2～3 … 21～22 22～23 23～0 Hourly flow 4.4％ 2.24％ 2.04％ … 5.28％ 0.58％ 0.12％ Total flow 4.4％ 6.64％ 8.68％ … 94.6％ 99.88％ 100％

[0143] In one example, the durations of multiple time periods are different and can be set according to specific circumstances. The historical traffic-time data can be:

[0144] Hour 0～1 1～3 3～5 … 21～22 22～23 23～0 Hourly flow 4.4％ 4.28％ 3.24％ … 5.28％ 5.58％ 4.12％ Total flow 4.4％ 8.68％ 11.92％ … 90.3％ 95.88％ 100％

[0145] In the embodiments of the present application, the traffic data is represented as a percentage of the total daily traffic. Therefore, the first progress requirements for multiple time periods can be equal to the traffic data of multiple time periods.

[0146] When the first progress requirements for multiple time periods are determined, the sub-exposure amounts corresponding to the time periods can be allocated according to the progress requirements.

[0147] In one example, when the time period is from 0 to 1 hour, the time period is 15 minutes, and this time period includes 4 time periods; then the sub-exposure amounts of the 4 time periods in this time period can evenly divide the first progress requirement.

[0148] For example, the target exposure volume of the multimedia resource is 10,000 times a day. According to the above second historical traffic-time data table, it can be determined that within the time period from 0 to 1 hour, the first progress requirement for exposure is 4.4%. Then, in the first time period, the number of exposures required is 440. Therefore, in the first time period, the sub-exposure amount for each time period is 110 times. That is, 110 times of exposure from 0 to 15 minutes, 110 times of exposure from 15 to 30 minutes, 110 times of exposure from 30 to 45 minutes, and 110 times of exposure from 45 to 60 minutes.

[0149] In the embodiments of the present application, by using the historical traffic-time data to determine the sub-exposure amounts corresponding to the time periods, the appropriate sub-exposure amounts can be adjusted according to the actual traffic distribution. So that in the time periods with less traffic, fewer sub-exposure amounts are allocated; in the time periods with more traffic, more sub-exposure amounts are allocated; to further ensure that the sub-exposure amounts for each time period can be completed, and the uniform exposure of the target exposure volume is achieved.

[0150] In one implementation manner, step S223 further includes:

[0151] Obtain the actual progress of the total historical time period; the total historical time period includes multiple time periods before the current time period;

[0152] Based on the actual progress and the first progress requirements of the total historical time period, obtain the first progress deviation;

[0153] Determine the second progress requirement for the current time period based on the first progress deviation and the first progress requirement for the current time period.

[0154] Update the sub-exposure amounts for multiple time segments in the current time period according to the second progress requirement.

[0155] The total historical time period is the multiple time periods that have passed from the start of executing the target volume of the multimedia resource to the current time. For example, according to the second historical traffic-time data table above, if the exposure of the multimedia resource has been executed for 5 hours, then the total historical time period includes 5 hours from 0 to 1, 1 to 3, and 3 to 5. The actual progress of the total historical time period is the exposure progress completed within the 5 hours from 0 to 1, 1 to 3, and 3 to 5.

[0156] In one example, the target volume of the multimedia resource is 10,000 exposures per day. According to the second historical traffic-time data table above, it can be determined that within the time period from 0 to 1 hour, the first progress requirement for exposure is 4.4%, so in the first time period, the number of exposures required is 440; within the time period from 1 to 3 hours, the first progress requirement for exposure is 4.28%, so in the second time period, the number of exposures required is 428. If the actual number of exposures completed from 0 hour to 1 hour is 400 times, then the progress is slow and there are still 40 times not completed; then in the second time period, adjust the first progress requirement so that the second time period can accelerate the exposure. The first progress requirement in the second time period can be adjusted to obtain the second progress requirement of 4.68%, that is, 468 exposures are required. Then, according to the second progress requirement of the time period, update the sub-exposure amounts for multiple time segments in the time period.

[0157] In one example, the target volume of the multimedia resource is 10,000 exposures per day. According to the second historical traffic-time data table above, it can be determined that within the time period from 0 to 1 hour, the first progress requirement for exposure is 4.4%, so in the first time period, the number of exposures required is 440; within the time period from 1 to 3 hours, the first progress requirement for exposure is 4.28%, so in the second time period, the number of exposures required is 428. If the actual number of exposures completed from 0 hour to 1 hour is 400 times, then the expected progress can be obtained as 4.4%, the actual progress is 4.0%, and the first progress deviation = (actual progress - expected progress) / expected progress = -0.091.

[0158] Based on the first progress deviation and the first progress requirement for the current time period, determine the second progress requirement for the current time period. When the first progress deviation exceeds the preset threshold range, the progress requirement for the current time period can be adjusted to determine the second progress requirement for the current time period. When the first progress deviation does not exceed the preset threshold range, no adjustment is required. For example, the preset threshold range can be (-0.05, 0.05).

[0159] In the above example, if the first progress deviation is -0.091, which exceeds the range (-0.05, 0.05), the adjustment coefficient can be calculated according to a preset formula, and then the first progress requirements for multiple subsequent time periods can be adjusted according to the adjustment coefficient. Exemplarily, the adjustment coefficient can be 1.8^(-2 * first progress deviation). The adjusted second progress requirement is the first progress requirement multiplied by the adjustment coefficient. Then, the sub-exposure amounts for multiple time periods are updated according to the second progress requirement.

[0160] In one example, the first progress requirements for multiple time periods can also be determined based on the first progress requirement for the time period, then the first progress deviation for the time period is calculated to obtain the adjustment coefficient, and then the first progress requirements for multiple subsequent time periods are multiplied by the adjustment coefficient respectively.

[0161] In the embodiments of the present application, by obtaining the first progress deviation based on the actual progress of the completed exposure and adjusting the exposure speed in a timely manner, it can be further ensured that the corresponding sub-exposure amounts are completed in each time period or time cycle, and uniform exposure of the target volume is achieved.

[0162] In one implementation manner, step S223 includes:

[0163] Obtain second historical data; the second historical data includes the first historical duration and the historical exposure amount within the first historical duration;

[0164] Determine the exposure weight of the multimedia resource according to the target volume and the second historical data;

[0165] Obtain the number of service requests in the previous time period of the current time period;

[0166] Determine the sub-exposure amount of the current time period according to the exposure weight and the number of service requests in the previous time period of the current time period.

[0167] In one example, if the total exposure amount in a day for the second historical data is 20,000 times and the target volume is to expose 10,000 times a day, then the exposure weight is 10,000 / 20,000 = 0.5.

[0168] In one example, the second historical data is that the total number of exposures in a day is 20,000 times, and the target approximate quantity is 30,000 exposures in three days. Then, the target exposure quantity for one day is determined to be 10,000 times, and the exposure weight is 10,000 / 20,000 = 0.5.

[0169] In one example, the second historical data is that the total number of exposures on the first day is 20,000 times, the total number of exposures on the second day is 40,000 times, and the total number of exposures on the third day is 25,000 times. The target approximate quantity is 30,000 exposures in three days. Then, the target approximate quantity for each day is determined to be 10,000 times. The exposure weight on the first day is 10,000 / 20,000 = 0.5; the exposure weight on the second day is 10,000 / 40,000 = 0.25; the exposure weight on the third day is 10,000 / 25,000 = 0.4.

[0170] In one example, the second historical data is that the total number of exposures on the first day is 20,000 times, the total number of exposures on the second day is 40,000 times, and the total number of exposures on the third day is 25,000 times. Then, the total number of exposures in three days is 85,000. The target approximate quantity is 30,000 exposures in three days. Then, the exposure weight for each day is determined to be 30,000 / 85,000 = 0.35%.

[0171] In the above examples, the target approximate quantity can also be unevenly distributed according to the second historical data.

[0172] In one example, the second historical data is that the total number of exposures in a day is 20,000 times, the target approximate quantity is 10,000 times in one day, and the exposure weight is 10,000 / 20,000 = 0.5. Then, the sub-approximate quantity for the first time period can be determined according to the average value. The time period is 15 minutes, and there are 96 time periods in a day. Then, the sub-exposure quantity for the first time period is 10,000 / 96 = 105 exposures. If the number of service requests in the first time period is 300 times, then the sub-exposure quantity for the second time period is 300*0.5 = 150 times.

[0173] In the embodiments of the present application, the previous time period of the current time period may be the time period before the current time period. For example, if the current time period is the third time period, then the previous time period of the current time period is the second time period.

[0174] In other embodiments of the present application, the total historical time period may be all the time periods before the current time period. For example, if the current time period is the fourth time period, then the total historical time period includes the first time period, the second time period, and the third time period.

[0175] In the embodiments of the present application, the exposure weight is determined through the second historical data and the target approximate quantity; then, the sub-exposure quantity of the current time period is determined according to the exposure weight and the number of service requests in the previous time period of the current time period.

[0176] In one embodiment, step S223 includes:

[0177] Obtain the orientation information of the multimedia resource;

[0178] Obtain third historical data, where the third historical data includes a second historical duration and a targeted exposure amount corresponding to the orientation information within the second historical duration;

[0179] Determine the exposure weight of the multimedia resource according to the target volume and the third historical data;

[0180] Obtain the number of targeted service requests in the previous time period of the current time period, where the number of targeted service requests is the number of service requests corresponding to the orientation information;

[0181] Determine the sub-exposure amount of the current time period according to the exposure weight and the number of targeted service requests in the previous time period of the current time period.

[0182] In one example, the orientation information of the multimedia resource may include user gender, user age, user region, etc.

[0183] In the embodiments of the present application, by obtaining the orientation information of the multimedia resource to perform targeted exposure on the multimedia resource, the click-through rate and conversion rate of the multimedia resource are further improved.

[0184] In one example, if the orientation information of the multimedia resource is male in Shanghai, then when the service request information is male in Shanghai, the multimedia resource is exposed in response to the service request.

[0185] In one example, the third historical data shows that the exposure volume in Shanghai in one day is 200,000 times, the exposure volume of males in Shanghai is 120,000 times, and the exposure volume of males over 18 years old in Shanghai is 64,000 times. When the orientation information of the multimedia resource received is male in Shanghai and the target volume is 10,000 exposures in one day, the exposure weight is \(10,000 / 120,000\).

[0186] In one example, the third historical data is the historical data of the previous cycle. For example, if the cycle is one week and today is Wednesday, the historical data of last Wednesday can be obtained as the third historical data.

[0187] In the embodiments of the present application, when the estimated traffic today is quite different from the traffic on the corresponding day of the previous cycle, the targeted exposure amount can be adjusted by a proportional coefficient. For example, today is Double Eleven and the corresponding day of the previous cycle is an ordinary day, then the traffic today is large, and the total exposure amount of the corresponding day of the previous cycle can be multiplied by a magnification factor to determine the exposure weight. Because the target volume today will also be larger. The magnification factor can be set according to human experience.

[0188] In one example, the amplification factor can also be determined based on the target volume for the current day and the target volume for the corresponding day in the previous cycle.

[0189] It can be understood that in some specific cases, a reduction factor can also be multiplied.

[0190] In one example, the second historical data can be the historical data of yesterday. In the case of a large difference in traffic, adjustment is made using the amplification factor or the reduction factor. For example, if yesterday was Friday and today is Saturday, the estimated traffic on Saturday is larger than that on Friday, and an amplification factor can be multiplied.

[0191] Then, based on the exposure weight and the number of targeted service requests in the previous time period of the current time period, the sub-exposure volume of the current time period is determined. The number of targeted service requests is the number of service requests from males in Shanghai. The method for determining the sub-exposure volume of the time period can adopt the implementation manner of any of the above aspects.

[0192] In the embodiment of the present application, by obtaining the targeted information of the multimedia resource and then determining the exposure weight based on the targeted exposure volume, the sub-exposure volume of the time period is determined, so that the multimedia resource can be targeted for exposure, improving the exposure effect of the multimedia resource.

[0193] In one implementation manner, there are multiple targeted information of the multimedia resource; step S223 further includes:

[0194] For each piece of targeted information, the sub-exposure volume corresponding to the time period is determined under the corresponding targeted information.

[0195] In one example, the targeted information of the multimedia resource includes males over 18 years old in Shanghai and males under 18 years old in Shanghai, the target volume is 6000 times for males over 18 years old in Shanghai and 4000 times for males under 18 years old in Shanghai. Then, according to the second historical data in the above example, the exposure weight for males over 18 years old in Shanghai is 6000 / 64000; the exposure weight for males under 18 years old in Shanghai is 4000 / (12000 - 64000). Then, the sub-exposure volume corresponding to the time period is calculated respectively for males over 18 years old in Shanghai and males under 18 years old in Shanghai.

[0196] In the embodiment of the present application, based on multiple different targeted information, the sub-exposure volume corresponding to the time period is determined respectively under the multiple targeted information, so that the target volume corresponding to the targeted information is completed respectively according to different targeted information, thereby realizing the targeted exposure of the multimedia resource.

[0197] In one implementation manner, step S223 further includes:

[0198] Obtain the completed exposure volume of the total historical time period;

[0199] Determine the completion weight of the total historical time period according to the target exposure amount and the completed exposure amount;

[0200] Obtain a second progress deviation according to the completion weight and the exposure weight of the total historical time period;

[0201] Obtain an adjustment coefficient according to a preset adjustment algorithm and the second progress deviation;

[0202] Update the sub-exposure amount of the current time period based on the adjustment coefficient.

[0203] In the embodiments of the present application, the completed exposure amount may refer to the number of times a multimedia resource is exposed within a corresponding time period. For example, the completed exposure amount of the total historical time period is the number of times the multimedia resource is exposed within the total historical time period.

[0204] In an example, the total exposure amount in a day for the second historical data is 20,000 times, the target amount is 10,000 times a day, and the exposure weight is 10,000 / 20,000 = 0.5; then the sub-exposure amount of the first time period can be determined according to the average value. The time period is 15 minutes, and a day includes 96 time periods, so the sub-exposure amount of the first time period is 10,000 / 96 = 105 exposures. If the request quantity in the first time period is 300 times, then the sub-exposure amount of the second time period is 300*0.5 = 150 times.

[0205] If 200 times have been exposed within the first time period, the actual progress is the exposed amount / total exposure amount, that is, 200 / 20,000 = 0.01. The expected progress is the exposure weight, that is, 0.5. The second progress deviation = (actual progress - expected progress) / expected progress = -0.98.

[0206] According to the second progress deviation, adjust the sub-exposure amount of the time period. It can be to adjust the first progress requirement of the time period when the second progress deviation exceeds the preset threshold range, and there is no need to adjust when the second progress deviation does not exceed the preset threshold range. For example, the preset threshold range can be (-0.05, 0.05).

[0207] In the above example, the second progress deviation is -0.98, which exceeds the range (-0.05, 0.05). Then, the adjustment coefficient can be calculated according to a preset formula, and the sub-exposure amounts of subsequent multiple time periods can be adjusted according to the adjustment coefficient. Exemplarily, the adjustment coefficient can be 1.8^(-2*second progress deviation). The adjusted sub-exposure amount is the original sub-exposure amount multiplied by the adjustment coefficient.

[0208] In the embodiments of the present application, by obtaining the second progress deviation according to the number of completed exposures and adjusting the exposure speed in a timely manner, it is possible to further ensure uniform exposure while completing the target amount.

[0209] In one embodiment, as Figure 6 shown, step S224 includes:

[0210] S2241, obtaining multiple true exposure volumes corresponding to multiple predicted click-through rate intervals in the previous time period of the current time period for the multimedia resource.

[0211] S2242, obtaining the maximum predicted exposure volume corresponding to multiple predicted click-through rate intervals based on the true exposure volumes; the maximum predicted exposure volume is the predicted exposure volume when the selection probability is 1.

[0212] S2243, obtaining the sum of multiple true exposure volumes corresponding to multiple predicted click-through rate intervals in the total historical time period for the multimedia resource, and the sum of multiple true click counts corresponding to multiple predicted click-through rate intervals. Wherein, the total historical time period includes multiple time periods before the current time period;

[0213] S2244, obtaining the average true click-through rate corresponding to multiple predicted click-through rate intervals based on the sum of multiple true exposure volumes and the sum of multiple true click counts.

[0214] S2245, obtaining the click-through rate target value for the multimedia resource in the current time period.

[0215] S2246, obtaining multiple selection probabilities corresponding to multiple predicted click-through rate intervals in the current time period based on the maximum predicted exposure volume, the average true click-through rate, the sub-exposure volume in the current time period, and the click-through rate target value corresponding to multiple predicted click-through rate intervals.

[0216] In one example, the exposure situation in the previous time period of the current time period is as shown in the following table:

[0217]

[0218]

[0219] Among them, the true exposure volume is obtained by statistical analysis based on the exposure situation. Then, the sum of multiple true click counts of multiple historical first time slices of multiple predicted click-through rate intervals is respectively counted; the sum of multiple true exposure volumes of multiple historical first time slices of multiple predicted click-through rate intervals is respectively counted; thus, for a specific predicted click-through rate interval, the average true CTR of the corresponding predicted click-through rate interval is calculated according to the sum of true click counts / the sum of true exposure volumes.

[0220] Based on the true exposure volume, assuming the selection probability is 100%, the maximum predicted exposure volume can be obtained.

[0221] The click-through rate target value can be set by staff or users, and is used to indicate the click-through rate expected to be obtained when the target volume is completed.

[0222] Based on the maximum estimated exposure volume corresponding to multiple estimated click-through rate intervals, the average true click-through rate, the sub-exposure volume in the current time period, and the click-through rate target value, the multiple selection probabilities corresponding to the multiple estimated click-through rate intervals in the current time period can be obtained as follows:

[0223] Construct the following system of equations (1):

[0224] cnt_1*p_1 + cnt_2*p_2 + … + cnt_50*p_50 = sub-volume;

[0225] (cnt_1*p_1*ctr_1 + cnt_2*p_2*ctr_2 + … + cnt_50*p_50*ctr_50) / sub-volume = click-through rate target value;

[0226] Solve the system of equations (1) to obtain p_1, p_2 … p_50, which are the multiple selection probabilities corresponding to the multiple estimated click-through rate intervals. Among them, the above system of equations (1) includes the known condition: 0 <= selection probability <= 1.

[0227] In the embodiments of the present application, based on the maximum estimated exposure volume corresponding to multiple estimated click-through rate intervals, the average true click-through rate, the sub-exposure volume in the current time period, and the click-through rate target value, the multiple selection probabilities corresponding to the multiple estimated click-through rate intervals in the current time period are obtained, so as to update the estimated click-through rate interval - selection probability comparison table in the current time period according to the exposed situation, dynamically adjust the exposure progress of the multimedia resource, and achieve uniform exposure of the target volume.

[0228] In one implementation manner, step S2245 includes:

[0229] Obtain a click-through rate random value according to the multiple average true click-through rates corresponding to multiple estimated click-through rate intervals and the maximum estimated exposure volume corresponding to the multiple estimated click-through rate intervals;

[0230] Determine the click-through rate random value as the click-through rate target value of the multimedia resource in the current time period.

[0231] In an example, click-through rate random value = (ctr_1*cnt_1 + ctr_2*cnt_2 + …) / (cnt_1 + cnt_2 + …).

[0232] The click-through rate random value can be a click-through rate value obtained based on the exposure situation, representing the random level of the click-through rate after exposure. By determining the click-through rate random value as the click-through rate target value, the click-through rate of the multimedia resource after exposure can conform to the random level. When there is no special requirement for the click-through rate, the click-through rate random value can be determined as the click-through rate target value without manual input.

[0233] In the embodiments of the present application, the click-through rate target value is determined through the average values of multiple true click-through rates and the maximum estimated exposure volume, and multiple selection probabilities are dynamically adjusted without the user inputting the click-through rate target value.

[0234] In one implementation manner, step S2245 includes:

[0235] Obtain the click-through rate random value according to the average values of multiple true click-through rates corresponding to multiple estimated click-through rate intervals and the maximum estimated exposure volume corresponding to multiple estimated click-through rate intervals;

[0236] Obtain the regulation coefficient;

[0237] Determine the click-through rate target value of the multimedia resource in the current time period according to the regulation coefficient and the click-through rate random value.

[0238] It can be understood that in order to increase the click-through rate of the multimedia resource, adjustment can be made by the regulation coefficient being greater than 1. For example, if it is desired that the click-through rate value is 130% higher than the current click-through rate value, the regulation coefficient can be set to 130%. So that the click-through rate target value = click-through rate random value * 130%.

[0239] In the embodiments of the present application, the higher the click-through rate target value, among the multiple selection probabilities obtained, the larger the selection probability tends to be in the estimated click-through rate interval with a larger estimated click-through rate value. Thus, with the same number of exposure times, the click-through rate is higher.

[0240] In the embodiments of the present application, the click-through rate target value is adjusted through the regulation coefficient, so that the click-through rate of the multimedia resource can be regulated.

[0241] In one implementation manner, step S224 further includes:

[0242] Determine the estimated click-through rate interval - selection probability comparison table of the multimedia resource in the first time period according to the first preset strategy; the first preset strategy includes: for multiple estimated click-through rate intervals, respectively set preset selection probability values, or respectively set selection probability random values.

[0243] The first time period can refer to the first time period starting from the execution of the target volume, the time period starting from 0s.

[0244] For multiple predicted click-through rate intervals, preset selection probability values are respectively set. Equal selection probability values can be set for multiple predicted click-through rate intervals. The equal selection probability value can be the target exposure volume divided by the total exposure volume of historical data. The historical data can be one-day historical data with little difference from the current-day traffic data.

[0245] In one example, if the target exposure volume is 1000 exposures in a day and the total exposure volume in a day of historical data is 20000 times, then the multiple selection probabilities for the first time period are 1000 / 20000.

[0246] For multiple predicted click-through rate intervals, preset selection probability values are respectively set, which can be artificially setting selection probability values for each predicted click-through rate interval.

[0247] For multiple predicted click-through rate intervals, random selection probability values are respectively set, which can be multiple random numbers obtained from the number of intervals corresponding to multiple predicted click-through rate intervals, and the multiple random numbers are corresponding to multiple predicted click-through rate intervals.

[0248] The embodiment of the present application determines the predicted click-through rate interval - selection probability comparison table of the multimedia resource in the first time period through the first preset strategy, so as to adjust the exposure of the multimedia resource in the first time period according to the predicted click-through rate interval - selection probability comparison table, and in subsequent time periods, update the predicted click-through rate interval - selection probability comparison table based on the exposed situation.

[0249] In one implementation manner, step S2246 includes: when the sum of the maximum predicted exposure volumes corresponding to multiple predicted click-through rate intervals is less than or equal to the sub-exposure volume of the current time period, determining that the multiple selection probabilities corresponding to the multiple predicted click-through rate intervals of the current time period are all 1.

[0250] When the sum of the maximum predicted exposure volumes corresponding to multiple predicted click-through rate intervals is less than or equal to the sub-exposure volume of the current time period, that is, when the selection probabilities are all 100%, and the sub-exposure volume still cannot be completed, then determine that each selection probability is 1, so as to complete the sub-exposure volume as quickly as possible, ensure the completion of the sub-exposure volume of the current time period, and further ensure the uniform exposure of the multimedia resource.

[0251] In one implementation manner, step S2246 includes: when the predicted click-through rate value obtained by exposing the sub-exposure volume of the current time period in the maximum predicted click-through rate interval is less than or equal to the click-through rate target value, determining that the multiple selection probabilities corresponding to the multiple predicted click-through rate intervals outside the maximum predicted click-through rate interval are all 0.

[0252] In one example, the sub-exposure amount for a time period is 100 exposures. Assuming that all 100 exposures are within the maximum estimated click-through rate range, for example, between 0.42 and 1, the estimated click-through rate can be obtained based on the average click-through rate within the maximum estimated click-through rate range. If this estimated click-through rate is less than the click-through rate target value, then to achieve the task of the click-through rate target value, the selection probabilities for the remaining estimated click-through rate ranges are all set to 0, so that the estimated click-through rate ranges during exposure all fall within the highest estimated click-through rate range. Subsequently, through progress adjustment, the sub-exposure amount is completed, thereby ensuring that the click-through rate target value is achieved.

[0253] The above situation can be expressed by the following formula:

[0254] cnt_1 + cnt_2 + … + cnt_i-1 + cnt_i * p_i = sub-exposure amount;

[0255] However, (cnt_1 * ctr_1 + cnt_2 * ctr_2 + … + cnt_i-1 * ctr_i-1 + cnt_i * p_i * ctr_i) / sub-exposure amount < click-through rate target value;

[0256] where 1 <= i <= 50 and p_i < 1.

[0257] In one example, the sub-exposure amount for a time period is 100 exposures. Assuming that all 100 exposures are within the maximum estimated click-through rate range, for example, between 0.42 and 1, the estimated click-through rate can be obtained based on the average click-through rate within the maximum estimated click-through rate range. This estimated click-through rate is greater than the click-through rate target value. However, the difference is extremely small. Then, to achieve the task of the click-through rate target value, the selection probabilities for the remaining estimated click-through rate ranges except for the first highest estimated click-through rate range and the second highest estimated click-through rate range can be all set to 0, so that the 100 exposures fall within the first highest estimated click-through rate range and the second highest estimated click-through rate range to ensure that the click-through rate target value is achieved.

[0258] In one implementation, when solving the system of equations (1) for multiple selection probabilities, the following method can be used for solving:

[0259] Example 1: Divide p_1 to p_50 into two equal parts:

[0260] p_1 = p_2 = … = p_i = p_up;

[0261] p_i+_1 = p_i+_2 = … = p_50 = p_down;

[0262] where 1 <= i <= 50;

[0263] Then the above system of equations (1) can be transformed into the following system of binary linear equations:

[0264] (cnt_1 + cnt_2 + … + cnt_i) * p_up + (cnt_i + 1 + cnt_I + 2 + … + cnt_50) * p_down = sub - exposure;

[0265] (cnt_1 * ctr_1 + cnt_2 * ctr_2 + … + cnt_i * ctr_i) * p_up + (cnt_i + 1 * ctr_i + 1 + cnt_r + 2 * ctr_i + 2 + … + cnt_50 * ctr_50) * p_down = click - through rate target value * sub - exposure;

[0266] Among them,

[0267] 0 <= p_up <= 1

[0268] 0 <= p_down <= 1

[0269] p_up >= p_down

[0270] Let i iterate one by one between 0 and 50 to solve for p_up and p_down that meet the constraints;

[0271] Example 2: Divide p_1 to p_50 into three equal - value parts:

[0272] p_1 = p_2 = … = p_i - 1 = 100%

[0273] pi = p_up

[0274] p_i + 1 = p_i + 2 = … = p_50 = p_down

[0275] where 1 <= i <= 50

[0276] Then the above - mentioned system of equations (1) can be transformed into the following system of binary linear equations:

[0277] (cnt_1 + cnt_2 + … + cnt_i - 1) + cnt_i * p_up + (cnt_i + 1 + cnt_i + 2 + … + cnt_50) * p_down = sub - exposure

[0278] (cnt_1 * ctr_1 + cnt_2 * ctr_2 + … + cnt_i - 1 * ctr_i - 1) + cnt_i * ctr_i * p_up + (cnt_i + 1 * ctr_i + 1 + cnt_r + 2 * ctr_i + 2 + … + cnt_50 * ctr_50) * p_down = click - through rate target value * sub - exposure

[0279] Among them,

[0280] 0 <= p_up <= 1

[0281] 0 <= p_down <= 1

[0282] p_up >= p_down

[0283] Similarly, iterate i one by one between 0 and 50 to solve for p_up and p_down that meet the constraints.

[0284] In one implementation, as Figure 7 shown, step S250 includes:

[0285] S251, when the first random factor is greater than or equal to the selection probability, determine the multimedia resource as the object to be exposed; to obtain multiple objects to be exposed;

[0286] S252, obtain the target number of exposed objects from the multiple objects to be exposed.

[0287] When a service request is received, the selection probabilities of multiple multimedia resources in the resource library can be determined. Thus, when the first random factor is obtained, determine the multiple multimedia resources with a selection probability less than the first random factor as the objects to be exposed. Then, determine the finally exposed multimedia resources from the multiple objects to be exposed.

[0288] In the embodiments of the present application, by first determining multiple objects to be exposed and then determining the finally exposed multimedia resources from the multiple objects to be exposed, it can be applicable to a multimedia resource system with a large number of resources, which is convenient for regulation.

[0289] In one implementation, step S253 includes:

[0290] Obtain the exposure progress of multiple objects to be exposed respectively;

[0291] Obtain a second random factor;

[0292] Based on the second random factor and the exposure progress, determine the exposed object.

[0293] In the embodiments of the present application, three random factors are involved: the first random factor, the second random factor, and the third random factor in the subsequent embodiments. Among them, the first random factor is a random number generated based on the first preset rule; the second random factor is a random number generated based on the second preset rule; the third random factor is a random number generated based on the third preset rule.

[0294] In one example, three multimedia resources are determined as objects to be exposed, and the situations are as follows: for the first one, the exposure progress is 0.3; for the second one, the exposure progress is 0.5; for the third one, the exposure progress is 0.6%. Then the exposure progress to be made is: for the first one, 0.7; for the second one, 0.5; for the third one, 0.4. Based on the exposure progress to be made, the exposure probability is determined as: for the first one, 0.7 / (0.7 + 0.5 + 0.4) = 7 / 16; for the second one, 0.5 / (0.7 + 0.5 + 0.4) = 5 / 16; for the third one, 0.4 / (0.7 + 0.5 + 0.4) = 4 / 16.

[0295] The second random factor is from 0 to 4 / 16, and the third exposure is determined; the second random factor is from 4 / 16 to 9 / 16, and the second exposure is determined; the second random factor is from 9 / 16 to 16 / 16, and the first exposure is determined.

[0296] In the embodiment of the present application, the multimedia resource to be exposed is determined according to the second random factor and the exposure progress, so that when determining the multimedia resource to be exposed, the progress factor is considered to ensure the uniform exposure of multiple multimedia resources.

[0297] In one implementation manner, step S253 includes:

[0298] Obtain the exposure weights of multiple objects to be exposed respectively;

[0299] Obtain the third random factor;

[0300] Determine the exposure object based on the third random factor and the exposure weight.

[0301] In one example, three multimedia resources are determined as objects to be exposed, and the situations are as follows: for the first one, the exposure weight is 0.3; for the second one, the exposure weight is 0.5; for the third one, the exposure weight is 0.7. Based on the exposure progress to be made, the exposure probability is determined as: for the first one, 0.3 / (0.7 + 0.5 + 0.3) = 3 / 15; for the second one, 0.5 / (0.7 + 0.5 + 0.3) = 5 / 15; for the third one, 0.7 / (0.7 + 0.5 + 0.3) = 7 / 15.

[0302] The third random factor is from 0 to 3 / 15, and the first exposure is determined; the third random factor is from 3 / 15 to 8 / 15, and the second exposure is determined; the third random factor is from 8 / 15 to 15 / 15, and the third exposure is determined.

[0303] In the embodiment of the present application, the multimedia resource to be exposed is determined according to the third random factor and the exposure weight, so that when determining the multimedia resource to be exposed, the exposure weight factor is considered to ensure the uniform exposure of multiple multimedia resources.

[0304] In the above implementation manner, the exposure weight can also be replaced by the selection probability.

[0305] The multimedia resource regulation method provided by the embodiment of the present application can dynamically adjust the exposure speed of multimedia resources, achieve uniform exposure under the target quantity, ensure the completion of the target quantity, and have good exposure effect.

[0306] The multimedia resource regulation method of the embodiment of the present application can also achieve directional exposure, meet the directional exposure requirements of multimedia resources; it can also meet the coordination between multiple multimedia resources under complex orientations without manual adjustment.

[0307] The multimedia resource regulation method of the embodiment of the present application can also regulate the click-through rate level to achieve exposure that meets the click-through rate target value.

[0308] Figure 8 The structural block diagram of a multimedia resource regulation device 800 according to an embodiment of the present application is shown. As Figure 8 shown, the multimedia resource regulation device 800 may include:

[0309] An estimated click-through rate interval determination module 810, configured to determine an estimated click-through rate interval of a multimedia resource;

[0310] A comparison table acquisition module 820, configured to acquire an estimated click-through rate interval - selection probability comparison table of the multimedia resource; wherein, the estimated click-through rate interval - selection probability comparison table includes the corresponding relationship between multiple estimated click-through rate intervals and multiple selection probabilities;

[0311] A selection probability determination module 830, configured to determine the selection probability corresponding to the estimated click-through rate interval according to the estimated click-through rate interval and the estimated click-through rate interval - selection probability comparison table;

[0312] A first random factor acquisition module 840, configured to acquire a first random factor; the first random factor is a random number generated according to a preset rule;

[0313] A regulation module 850, configured to regulate the exposure of the multimedia resource according to the first random factor and the selection probability.

[0314] In one implementation, the estimated click-through rate interval determination module 810 includes:

[0315] An interval set acquisition sub-module, configured to acquire an interval set of the multimedia resource, and the interval set includes multiple estimated click-through rate intervals divided according to a preset rule;

[0316] A target estimated click-through rate acquisition sub-module, configured to acquire the target estimated click-through rate of the multimedia resource;

[0317] An estimated click-through rate interval determination sub-module, configured to determine the estimated click-through rate interval in which the target estimated click-through rate falls as the estimated click-through rate interval of the multimedia resource.

[0318] In one embodiment, the interval set obtaining sub-module includes:

[0319] The first historical data obtaining sub-module is configured to obtain first historical data, where the first historical data includes historical estimated click-through rate and historical exposure volume corresponding to the historical estimated click-through rate;

[0320] The interval set obtaining sub-module is configured to determine multiple estimated click-through rate intervals of the multimedia resource based on the historical estimated click-through rate and the historical exposure volume, where the historical exposure volumes of the multiple estimated click-through rate intervals are equal.

[0321] In one embodiment, the target estimated click-through rate obtaining sub-module includes:

[0322] The first estimated click-through rate obtaining sub-module is configured to determine a first estimated click-through rate of the multimedia resource according to the content information of the multimedia resource;

[0323] The second estimated click-through rate obtaining sub-module is configured to determine a second estimated click-through rate of the multimedia resource according to the service request information;

[0324] The target estimated click-through rate obtaining sub-module is configured to obtain a target estimated click-through rate according to the first estimated click-through rate and / or the second estimated click-through rate.

[0325] In one embodiment, the look-up table obtaining module 820 is configured to generate a look-up table of estimated click-through rate interval - selection probability of the multimedia resource based on the remaining volume of the multimedia resource at an interval of a first preset time; where the remaining volume includes remaining exposure duration and remaining exposure volume.

[0326] In one embodiment, the look-up table obtaining module 820 includes:

[0327] The target volume obtaining sub-module is configured to obtain a target volume of the multimedia resource, where the target volume includes a target duration and a target exposure volume;

[0328] The time period dividing sub-module is configured to divide the target duration into multiple time periods;

[0329] The sub-exposure volume determining sub-module is configured to determine a sub-exposure volume corresponding to the time period according to the target exposure volume;

[0330] The look-up table determining sub-module is configured to determine a look-up table of estimated click-through rate interval - selection probability corresponding to the time period based on the sub-exposure volume corresponding to the time period.

[0331] In one embodiment, the sub-exposure volume determining sub-module is configured to:

[0332] Obtain historical traffic-time data, where the historical traffic-time data includes multiple time periods and the traffic corresponding to each of the multiple time periods;

[0333] Based on the historical traffic-time data, determine the first progress requirement for multiple time periods; the first progress requirement is the progress of exposure to be completed for the corresponding time period;

[0334] Based on the target exposure volume and the first progress requirement of the time period, determine the sub-exposure volumes for multiple time segments.

[0335] In one implementation, the sub-exposure volume determination sub-module is further configured to:

[0336] Obtain the actual progress of the total historical time period; the total historical time period includes multiple time periods before the current time period;

[0337] Based on the actual progress and the first progress requirement of the total historical time period, obtain the first progress deviation;

[0338] Based on the first progress deviation and the first progress requirement of the current time period, determine the second progress requirement of the current time period;

[0339] According to the second progress requirement, update the sub-exposure volumes corresponding to multiple time segments in the current time period.

[0340] In one implementation, the sub-exposure volume determination sub-module is configured to:

[0341] Obtain second historical data; the second historical data includes a first historical duration and the historical exposure volume within the first historical duration;

[0342] According to the target volume and the second historical data, determine the exposure weight of the multimedia resource;

[0343] Obtain the number of business requests in the previous time segment of the current time segment;

[0344] According to the exposure weight and the number of business requests in the previous time segment of the current time segment, determine the sub-exposure volume of the current time segment.

[0345] In one implementation, the sub-exposure volume determination sub-module is configured to:

[0346] Obtain the targeting information of the multimedia resource;

[0347] Obtain third historical data, where the third historical data includes a second historical duration and the targeted exposure volume corresponding to the targeting information within the second historical duration;

[0348] According to the target volume and the third historical data, determine the exposure weight of the multimedia resource;

[0349] Obtain the number of targeted service requests in the previous time period of the current time period, where the number of targeted service requests is the number of service requests corresponding to the corresponding targeted information;

[0350] Determine the sub-exposure volume of the current time period according to the exposure weight and the number of targeted service requests in the previous time period of the current time period.

[0351] In one implementation, there are multiple targeted information of the multimedia resource;

[0352] The sub-exposure volume determination sub-module is further used for:

[0353] For each targeted information, determine the sub-exposure volume of the time period corresponding to the corresponding targeted information.

[0354] In one implementation, the sub-exposure volume determination sub-module is further used for:

[0355] Obtain the completed exposure volume of the total historical time period; the total historical time period includes multiple time periods before the current time period;

[0356] Determine the completion weight of the total historical time period according to the target exposure volume and the completed exposure volume;

[0357] Obtain the second progress deviation according to the completion weight and the exposure weight of the total historical time period;

[0358] Obtain the adjustment coefficient according to the preset adjustment algorithm and the second progress deviation;

[0359] Update the sub-exposure volume of the current time period based on the adjustment coefficient.

[0360] In one implementation, the look-up table determination sub-module includes:

[0361] The true exposure volume acquisition sub-module is used to obtain multiple true exposure volumes corresponding to multiple estimated click-through rate intervals in the previous time period of the current time period for the multimedia resource;

[0362] The maximum estimated exposure volume acquisition sub-module is used to obtain the maximum estimated exposure volume corresponding to multiple estimated click-through rate intervals based on multiple true exposure volumes; among them, the maximum estimated exposure volume is the estimated exposure volume when the selection probability is 1;

[0363] The true click-through rate average acquisition sub-module is used to obtain the sum of multiple true exposure volumes corresponding to multiple estimated click-through rate intervals and the sum of multiple true click numbers corresponding to multiple estimated click-through rate intervals for the multimedia resource in the total historical time period; the total historical time period includes multiple time periods before the current time period; based on the sum of multiple true exposure volumes and the sum of multiple true click numbers, obtain the true click-through rate average corresponding to multiple estimated click-through rate intervals;

[0364] A click-through rate target value acquisition sub-module, which is used to acquire the click-through rate target value of the multimedia resource in the current time period;

[0365] A selection probability determination sub-module, which is used to obtain multiple selection probabilities corresponding to multiple estimated click-through rate intervals in the current time period based on the maximum estimated exposure volume, the average true click-through rate, the sub-exposure volume in the current time period, and the click-through rate target value corresponding to the multiple estimated click-through rate intervals.

[0366] In one implementation manner, the click-through rate target value acquisition sub-module is used to:

[0367] Obtain a click-through rate random value according to the multiple average true click-through rates corresponding to the multiple estimated click-through rate intervals and the maximum estimated exposure volume corresponding to the multiple estimated click-through rate intervals;

[0368] Determine the click-through rate random value as the click-through rate target value of the multimedia resource in the current time period.

[0369] In one implementation manner, the click-through rate target value acquisition sub-module is used to:

[0370] Obtain a click-through rate random value according to the multiple average true click-through rates corresponding to the multiple estimated click-through rate intervals and the maximum estimated exposure volume corresponding to the multiple estimated click-through rate intervals;

[0371] Obtain a regulation coefficient;

[0372] Determine the click-through rate target value of the multimedia resource in the current time period according to the regulation coefficient and the click-through rate random value.

[0373] In one implementation manner, the look-up table determination sub-module is further used to:

[0374] Determine a look-up table of estimated click-through rate intervals - selection probabilities of the multimedia resource in the first time period according to a first preset strategy; the first preset strategy includes: respectively setting preset selection probability values or respectively setting selection probability random values for the multiple estimated click-through rate intervals.

[0375] In one implementation manner, the selection probability determination sub-module is used to:

[0376] In the case where the sum of the maximum estimated exposure volumes corresponding to the multiple estimated click-through rate intervals is less than or equal to the sub-exposure volume in the current time period, determine that the multiple selection probabilities corresponding to the multiple estimated click-through rate intervals in the current time period are all 1.

[0377] In one implementation manner, the selection probability determination sub-module is used to:

[0378] When the sub-exposure amounts in the current time period are all exposed within the maximum estimated click-through rate range and the estimated click-through rate values obtained are less than or equal to the click-through rate target value, it is determined that the selection probabilities corresponding to multiple estimated click-through rate ranges outside the maximum estimated click-through rate range are all 0.

[0379] In one implementation, the regulation module 850 includes:

[0380] The to-be-exposed object acquisition sub-module is used to determine the multimedia resource as the to-be-exposed object when the first random factor is greater than or equal to the selection probability, so as to obtain multiple to-be-exposed objects;

[0381] The exposure sub-module is used to obtain a target number of multimedia resources as exposure objects from multiple to-be-exposed objects.

[0382] In one implementation, the exposure sub-module includes:

[0383] The exposure progress acquisition sub-module is used to respectively acquire the exposure progress of multiple to-be-exposed objects;

[0384] The second random factor acquisition sub-module is used to acquire the second random factor;

[0385] The exposure object determination sub-module is used to determine the exposure object based on the second random factor and the exposure progress.

[0386] In one implementation, the exposure sub-module includes:

[0387] The exposure weight acquisition sub-module is used to respectively acquire the exposure weights of multiple to-be-exposed objects;

[0388] The third random factor acquisition sub-module is used to acquire the third random factor;

[0389] The exposure object determination sub-module is used to determine the exposure object based on the third random factor and the exposure weight.

[0390] For the functions of the modules in each device of the embodiments of the present application, reference may be made to the corresponding descriptions in the above-mentioned multimedia resource regulation method, which will not be elaborated herein.

[0391] Figure 9 Schematically shows a hardware architecture diagram of a computer device 900 suitable for implementing the multimedia resource regulation method according to Embodiment 3 of the present application. In this embodiment, the computer device 900 is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. For example, it can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a rack-mounted server, a blade server, a tower server or a cabinet server (including an independent server or a server cluster composed of multiple servers), etc. As Figure 9As shown, the computer device 900 at least includes, but is not limited to: a memory 910, a processor 920, and a network interface 930 that can communicate with each other through a system bus. Among them:

[0392] The memory 910 at least includes one type of computer-readable storage medium. The readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disc, etc. In some embodiments, the memory 910 can be an internal storage module of the computer device 900, such as the hard disk or memory of the computer device 900. In other embodiments, the memory 910 can also be an external storage device of the computer device 900, such as a plug-in hard disk equipped on the computer device 900, a Smart Media Card (SMC for short), a Secure Digital (SD for short) card, a Flash Card, etc. Of course, the memory 910 can also include both the internal storage module and the external storage device of the computer device 900. In this embodiment, the memory 910 is generally used to store the operating system and various application software installed on the computer device 900, such as the program code of the multimedia resource regulation method. In addition, the memory 910 can also be used to temporarily store various data that have been output or will be output.

[0393] In some embodiments, the processor 920 can be a central processing unit (CPU for short), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 920 is generally used to control the overall operation of the computer device 900, such as performing control and processing related to data interaction or communication with the computer device 900. In this embodiment, the processor 920 is used to run the program code stored in the memory 910 or process data.

[0394] The network interface 930 may include a wireless network interface or a wired network interface, which is generally used to establish a communication link between the computer device 900 and other computer devices. For example, the network interface 930 is used to connect the computer device 900 to an external terminal via a network, and to establish a data transmission channel and a communication link between the computer device 900 and the external terminal. The network may be a wireless or wired network such as an enterprise intranet (Intranet), the Internet, the Global System of Mobile communication (GSM for short), Wideband Code Division Multiple Access (WCDMA for short), 4G network, 9G network, Bluetooth, Wi-Fi, etc.

[0395] It should be noted that Figure 9 Only the computer device with components 910-930 is shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively.

[0396] In this embodiment, the multimedia resource control method stored in the memory 910 can also be divided into one or more program modules and executed by one or more processors (processor 920 in this embodiment) to complete the embodiments of the present application.

[0397] Embodiment 4

[0398] The embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the multimedia resource control method in the embodiments are implemented.

[0399] In this embodiment, the computer-readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the computer-readable storage medium may be an internal storage unit of a computer device, such as the hard disk or memory of the computer device. In other embodiments, the computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk equipped on the computer device, a Smart Media Card (abbreviated as SMC), a Secure Digital (abbreviated as SD) card, a Flash Card, etc. Of course, the computer-readable storage medium may also include both the internal storage unit and the external storage device of the computer device. In this embodiment, the computer-readable storage medium is generally used to store the operating system and various application software installed on the computer device, such as the program code of the multimedia resource control method in the embodiment. In addition, the computer-readable storage medium can also be used to temporarily store various data that have been output or will be output.

[0400] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the embodiments of the present application can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device, so that they can be stored in the storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order than here, or they can be separately made into individual integrated circuit modules, or multiple modules or steps among them can be made into a single integrated circuit module to implement. In this way, the embodiments of the present application are not limited to any specific combination of hardware and software.

[0401] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A multimedia resource regulation method, characterized in that, Including: Determine the estimated click-through rate range of the multimedia resource; Obtain the estimated click-through rate range - selection probability comparison table of the multimedia resource; wherein, the estimated click-through rate range - selection probability comparison table includes the corresponding relationship between multiple estimated click-through rate ranges and multiple selection probabilities; Determine the selection probability corresponding to the estimated click-through rate range according to the estimated click-through rate range and the estimated click-through rate range - selection probability comparison table; Obtain a first random factor, where the first random factor is a random number generated according to a preset rule; Regulate the exposure of the multimedia resource according to the first random factor and the selection probability; Wherein, the obtaining of the estimated click-through rate range - selection probability comparison table of the multimedia resource includes: Obtain the target volume of the multimedia resource, where the target volume includes a target duration and a target exposure volume; Divide the target duration into multiple time periods; Determine the sub-exposure volume corresponding to the time period according to the target exposure volume; Based on the sub-exposure volume corresponding to the time period, determine the estimated click-through rate range - selection probability comparison table corresponding to the time period; Wherein, the determining of the sub-exposure volume corresponding to the time period according to the target exposure volume further includes: Obtain the actual progress of the total historical time period, where the total historical time period includes multiple time periods before the current time period; Obtain a first progress deviation based on the actual progress and the first progress requirement of the total historical time period; Determine the second progress requirement of the current time period based on the first progress deviation and the first progress requirement of the current time period; Update the sub-exposure volumes corresponding to multiple time periods in the current time period according to the second progress requirement.

2. The method according to claim 1, wherein The obtaining of the estimated click-through rate range of the multimedia resource includes: Obtain the range set of the multimedia resource, where the range set includes multiple estimated click-through rate ranges divided according to a preset rule; Obtain the target estimated click-through rate of the multimedia resource; Determine the estimated click-through rate range into which the target estimated click-through rate falls as the estimated click-through rate range of the multimedia resource.

3. The method according to claim 2, wherein The obtaining of the range set of the multimedia resource includes: Obtain first historical data, where the first historical data includes historical estimated click-through rates and the historical exposure volumes corresponding to the historical estimated click-through rates; Based on the historical estimated click-through rates and the historical exposure volumes, determine multiple estimated click-through rate ranges of the multimedia resource, where the historical exposure volumes of the multiple estimated click-through rate ranges are equal.

4. The method according to claim 2, wherein The obtaining of the target estimated click-through rate of the multimedia resource includes: Determine the first estimated click-through rate of the multimedia resource according to the content information of the multimedia resource; Determine the second estimated click-through rate of the multimedia resource according to the service request information; Obtain the target estimated click-through rate according to the first estimated click-through rate and / or the second estimated click-through rate.

5. The method according to claim 1, wherein The obtaining of the estimated click-through rate range - selection probability comparison table of the multimedia resource includes: At intervals of a first preset time, generate a predicted click-through rate interval - selection probability comparison table for the multimedia resource based on the remaining approximate amount of the multimedia resource; wherein the remaining approximate amount includes the remaining exposure duration and the remaining exposure amount.

6. The method according to claim 1, wherein The determining of the sub-exposure amount corresponding to the time period according to the target exposure amount includes: Obtain historical traffic-time data, where the historical traffic-time data includes multiple time periods and the traffic corresponding to each of the multiple time periods; Based on the multiple time periods and the traffic corresponding to each of the multiple time periods, determine the first progress requirements for the multiple time periods, where the first progress requirements are the progress of exposure that needs to be completed for the corresponding time periods; Based on the target exposure amount and the first progress requirements, determine the sub-exposure amounts corresponding to multiple time periods.

7. The method according to claim 1, characterized in that The determining of the sub-exposure amount corresponding to the time period according to the target exposure amount includes: Obtain second historical data; the second historical data includes a first historical duration and the historical exposure amount within the first historical duration; According to the target approximate amount and the second historical data, determine the exposure weight of the multimedia resource; Obtain the number of service requests in the previous time period of the current time period; According to the exposure weight and the number of service requests in the previous time period, determine the sub-exposure amount of the current time period.

8. The method according to claim 1, characterized in that, The determining of the sub-exposure amount corresponding to the time period according to the target exposure amount includes: Obtain the targeting information of the multimedia resource; Obtain third historical data, where the third historical data includes a second historical duration and the targeted exposure amount corresponding to the targeting information within the second historical duration; According to the target approximate amount and the third historical data, determine the exposure weight of the multimedia resource; Obtain the number of targeted service requests in the previous time period of the current time period, where the number of targeted service requests is the number of service requests corresponding to the targeting information; According to the exposure weight and the number of targeted service requests in the previous time period, determine the sub-exposure amount of the current time period.

9. The method according to claim 8, wherein There are multiple targeting information for the multimedia resource; The determining of the sub-exposure amount corresponding to the time period according to the target exposure amount further includes: For each targeting information, determine the sub-exposure amount corresponding to the time period under the corresponding targeting information.

10. The method according to any one of claims 7 to 9, characterized in that, The determining of the sub-exposure amount corresponding to the time period according to the target exposure amount further includes: Obtain the completed exposure amount of the total historical time period, where the total historical time period includes multiple time periods before the current time period; According to the target exposure amount and the completed exposure amount, determine the completion weight of the total historical time period; According to the completion weight and the exposure weight of the total historical time period, obtain a second progress deviation; According to a preset adjustment algorithm and the second progress deviation, obtain an adjustment coefficient; Based on the adjustment coefficient, update the sub-exposure amount of the current time period.

11. The method according to claim 1, characterized in that, The determining of the predicted click-through rate interval - selection probability comparison table corresponding to the time period based on the sub-exposure amount corresponding to the time period includes: Obtain the multiple actual exposure amounts corresponding to multiple predicted click-through rate intervals for the multimedia resource in the previous time period of the current time period; Based on the multiple true exposure volumes, obtain the maximum estimated exposure volume corresponding to multiple estimated click-through rate intervals; wherein, the maximum estimated exposure volume is the estimated exposure volume when the selection probability is 1. Obtain the total sum of multiple true exposure volumes corresponding to multiple estimated click-through rate intervals and the total sum of multiple true click numbers corresponding to multiple estimated click-through rate intervals for the multimedia resource in the total historical time period; wherein, the total historical time period includes multiple time periods before the current time period. Based on the total sum of the multiple true exposure volumes and the total sum of the multiple true click numbers, obtain the average true click-through rate corresponding to multiple estimated click-through rate intervals. Obtain the click-through rate target value of the multimedia resource in the current time period. Based on the maximum estimated exposure volume, the average true click-through rate, the sub-exposure volume in the current time period, and the click-through rate target value corresponding to multiple estimated click-through rate intervals, obtain multiple selection probabilities corresponding to multiple estimated click-through rate intervals in the current time period.

12. The method according to claim 11, wherein The obtaining the click-through rate target value of the multimedia resource in the current time period includes: Obtain a click-through rate random value according to the average true click-through rate corresponding to multiple estimated click-through rate intervals and the maximum estimated exposure volume corresponding to multiple estimated click-through rate intervals. Determine the click-through rate random value as the click-through rate target value of the multimedia resource in the current time period.

13. The method according to claim 11, wherein The obtaining the click-through rate target value of the multimedia resource in the current time period includes: Obtain a click-through rate random value according to the average true click-through rate corresponding to multiple estimated click-through rate intervals and the maximum estimated exposure volume corresponding to multiple estimated click-through rate intervals. Obtain a regulation coefficient. Determine the click-through rate target value of the multimedia resource in the current time period according to the regulation coefficient and the click-through rate random value.

14. The method according to claim 11, wherein The determining the estimated click-through rate interval - selection probability comparison table for the time period based on the sub-exposure volume further includes: Determine the estimated click-through rate interval - selection probability comparison table for the multimedia resource in the first time period according to a first preset strategy; the first preset strategy includes: respectively setting preset selection probability values or respectively setting selection probability random values corresponding to multiple estimated click-through rate intervals.

15. The method according to claim 11, wherein The obtaining multiple selection probabilities corresponding to multiple estimated click-through rate intervals in the current time period based on the maximum estimated exposure volume, the average true click-through rate, the sub-exposure volume in the current time period, and the click-through rate target value corresponding to multiple estimated click-through rate intervals includes: In the case where the total sum of the maximum estimated exposure volumes corresponding to multiple estimated click-through rate intervals is less than or equal to the sub-exposure volume in the current time period, determine that the multiple selection probabilities corresponding to multiple estimated click-through rate intervals in the current time period are all 1.

16. The method according to claim 11, wherein The obtaining multiple selection probabilities corresponding to multiple estimated click-through rate intervals in the current time period based on the maximum estimated exposure volume, the average true click-through rate, the sub-exposure volume in the current time period, and the click-through rate target value corresponding to multiple estimated click-through rate intervals includes: When the sub-exposure amounts in the current time period are all exposed within the maximum estimated click-through rate range and the estimated click-through rate value obtained is less than or equal to the click-through rate target value, it is determined that the selection probabilities corresponding to multiple estimated click-through rate ranges outside the maximum estimated click-through rate range are all 0.

17. The method according to claim 1, wherein The adjusting the exposure of the multimedia resource according to the first random factor and the selection probability includes: When the first random factor is greater than or equal to the selection probability, determining that the multimedia resource is an object to be exposed to obtain multiple objects to be exposed; Obtaining a target number of exposed objects from the multiple objects to be exposed.

18. The method according to claim 17, wherein The obtaining a target number of exposed objects from the multiple objects to be exposed includes: Respectively obtaining the exposure progress of the multiple objects to be exposed; Obtaining a second random factor; Determining the exposed object based on the second random factor and the exposure progress.

19. The method according to claim 17, characterized in that, The obtaining a target number of exposed objects from the multiple objects to be exposed includes: Respectively obtaining the exposure weights of the multiple objects to be exposed; Obtaining a third random factor; Determining the exposed object based on the third random factor and the exposure weight.

20. A multimedia resource control device, characterized in that, including: An estimated click-through rate range determining module, configured to determine the estimated click-through rate range of the multimedia resource; A look-up table obtaining module, configured to obtain an estimated click-through rate range - selection probability look-up table of the multimedia resource; wherein, the estimated click-through rate range - selection probability look-up table includes the corresponding relationship between multiple estimated click-through rate ranges and multiple selection probabilities; A selection probability determining module, configured to determine the selection probability corresponding to the estimated click-through rate range according to the estimated click-through rate range - selection probability look-up table; A first random factor obtaining module, configured to obtain a first random factor; the first random factor is a random number generated according to a preset rule; An adjusting module, adjusting the exposure of the multimedia resource according to the first random factor and the selection probability; Wherein, the look-up table obtaining module is further configured to: Obtain the target volume of the multimedia resource, and the target volume includes a target duration and a target exposure amount; Dividing the target duration into multiple time periods; Determining the sub-exposure amount corresponding to the time period according to the target exposure amount; Based on the sub-exposure amount corresponding to the time period, determining the estimated click-through rate range - selection probability look-up table corresponding to the time period; Wherein, the look-up table obtaining module is further configured to: Obtain the real progress of the total historical time period, and the total historical time period includes multiple time periods before the current time period; Obtaining a first progress deviation based on the real progress and the first progress requirement of the total historical time period; Determining the second progress requirement of the current time period based on the first progress deviation and the first progress requirement of the current time period; Updating the sub-exposure amounts corresponding to multiple time periods in the current time period according to the second progress requirement.

21. An electronic device, characterized in that, including: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method according to any one of claims 1-19.

22. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method according to any one of claims 1-19 is implemented.

23. A computer program product, the computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by a processor, the steps of the method according to any one of claims 1-19 are implemented.

Images