Method and device for determining fostering flow and related equipment

Abstract

The invention provides a fostering traffic determination method and device and related equipment, and the method comprises the steps: obtaining the distribution efficiency index information of each candidate fostering video in a plurality of candidate fostering videos, the distribution efficiency index information is an index value used for representing the distribution efficiency of the candidate fostering videos in a preset historical time period; screening from the plurality of candidate fostering videos based on the distribution efficiency index information to obtain at least one to-be-fostering video; based on the distribution efficiency index information of the to-be-foster videos and pre-acquired to-be-foster total traffic, the foster traffic of the to-be-foster videos is determined, the to-be-foster total traffic is the sum of the foster traffic of the at least one to-be-foster video, and the foster traffic is the total foster traffic of the at least one to-be-foster video when the to-be-foster video is popularized in the current time window. The to-be-fostering video distributes the obtained flow. In this way, the flexibility of the method for determining the fostering flow of the to-be-fostering video is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of big data technology, and in particular, to a method and device for determining support traffic and related equipment. Background Art

[0002] With the continuous development of communication technology, watching videos has played an increasingly important role in people's lives. In the actual operation process, in order to support some videos to be supported, it is usually necessary to first determine the support traffic of the videos to be supported. Currently, when determining the support traffic of the videos to be supported, the support traffic of the videos to be supported is usually allocated in a fixed proportion, resulting in poor flexibility in the way of determining the support traffic of the videos to be supported. Summary of the Invention

[0003] The embodiments of the present application provide a method and device for determining support traffic and related equipment to solve the problem of poor flexibility in the way of determining the support traffic of the videos to be supported.

[0004] To solve the above problems, the present application is implemented as follows:

[0005] In a first aspect, the embodiments of the present application provide a method for determining support traffic, including:

[0006] Obtain the distribution efficiency index information of each candidate support video among a plurality of candidate support videos, where the distribution efficiency index information is an index value used to represent the distribution efficiency of the candidate support video within a preset historical time period, and the preset historical time period includes the time period before the current time window;

[0007] Based on the distribution efficiency index information, screen at least one video to be supported from the plurality of candidate support videos;

[0008] Based on the distribution efficiency index information of the video to be supported and the pre-obtained total support traffic, determine the support traffic of the video to be supported, where the total support traffic is the sum of the support traffic of the at least one video to be supported, and the support traffic is the traffic allocated to the video to be supported when promoting the video to be supported in the current time window.

[0009] In a second aspect, the embodiments of the present application provide a device for determining support traffic, including:

[0010] A first acquisition module, configured to obtain the distribution efficiency index information of each candidate support video among a plurality of candidate support videos, where the distribution efficiency index information is an index value used to represent the distribution efficiency of the candidate support video within a preset historical time period, and the preset historical time period includes the time period before the current time window;

[0011] A screening module, configured to screen at least one video to be supported from the multiple candidate supported videos based on the distribution efficiency index information;

[0012] A first determination module, configured to determine the support traffic of the video to be supported based on the distribution efficiency index information of the video to be supported and the total support traffic obtained in advance, where the total support traffic is the sum of the support traffic of the at least one video to be supported, and the support traffic is the traffic allocated to the video to be supported when promoting the video to be supported in the current time window.

[0013] In a third aspect, an embodiment of the present application further provides an electronic device, including: a memory, a processor, and a program stored on the memory and executable on the processor; the processor is configured to read the program in the memory to implement the steps in the method described in the first aspect above.

[0014] In a fourth aspect, an embodiment of the present application further provides a readable storage medium for storing a program, where the program, when executed by a processor, implements the steps in the method described in the first aspect above.

[0015] In a fifth aspect, an embodiment of the present application further provides a computer program product, including computer instructions, where the computer instructions, when executed by a processor, implement the steps in the method described in the first aspect above.

[0016] In an embodiment of the present application, the distribution efficiency index information of each candidate supported video in the multiple candidate supported videos is obtained, where the distribution efficiency index information is an index value used to represent the distribution efficiency of the candidate supported video in a preset historical time period, and the preset historical time period includes the time period before the current time window; at least one video to be supported is screened from the multiple candidate supported videos based on the distribution efficiency index information; the support traffic of the video to be supported is determined based on the distribution efficiency index information of the video to be supported and the total support traffic obtained in advance, where the total support traffic is the sum of the support traffic of the at least one video to be supported, and the support traffic is the traffic allocated to the video to be supported when promoting the video to be supported in the current time window. In this way, the support traffic of the video to be supported can be flexibly determined based on the distribution efficiency index information of the video to be supported and the total support traffic obtained in advance, thereby improving the flexibility of the determination method of the support traffic of the video to be supported. Description of the Drawings

[0017] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments of the present application. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0018] Figure 1 is one of the flow diagrams of the method for determining supported traffic provided by the embodiments of the present application;

[0019] Figure 2 is a schematic diagram of the user preference information table provided by the embodiments of the present application;

[0020] Figure 3 is the second of the flow diagrams of the method for determining supported traffic provided by the embodiments of the present application;

[0021] Figure 4 is a schematic diagram of the structure of the device for determining supported traffic provided by the embodiments of the present application;

[0022] Figure 5 is a schematic diagram of the structure of the electronic device provided by the embodiments of the present application. Detailed implementation manners

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present application belong to the scope of protection of the present application.

[0024] Terms such as "first" and "second" in the embodiments of the present application are used to distinguish similar objects and do not necessarily describe a specific order or sequence. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices. In addition, in the present application, "and / or" is used to represent at least one of the connected objects. For example, A and / or B and / or C represents seven situations including A alone, B alone, C alone, A and B both present, B and C both present, A and C both present, and A, B, and C all present.

[0025] Please refer to Figure 1 , Figure 1 which is a flow diagram of the method for determining supported traffic provided by the embodiments of the present application. Figure 1The supported traffic determination method shown can be executed by an electronic device.

[0026] As Figure 1 shown, the supported traffic determination method may include the following steps:

[0027] Step 101: Obtain the distribution efficiency index information of each candidate supported video among multiple candidate supported videos. The distribution efficiency index information is an index value used to represent the distribution efficiency of the candidate supported video within a preset historical time period, and the preset historical time period includes the time period before the current time window.

[0028] Among them, the multiple candidate supported videos can be the videos in the supported video content pool, and the supported video content pool can be a database in the electronic device of the embodiment of the present application, or the supported video content pool can also be a database outside the electronic device of the embodiment of the present application, that is, the supported video content pool can also be a database in other electronic devices, and the other electronic devices can be communicatively connected to the electronic device in the embodiment of the present application.

[0029] Among them, the specific type of the candidate supported video is not limited here. Optionally, the candidate supported video can be a new video, and the new video can be understood as a video with an online duration less than or equal to 30 days; further optionally, the candidate supported video can be a new video in the hot broadcast period and determined to require traffic support. The above hot broadcast period can be understood as: if the number of visits to the new video exceeds a preset visit value, it can be determined that the new video is in the hot broadcast period. And determining whether traffic support is required can be determined by the user or automatically determined by the electronic device in the embodiment of the present application. For example, when the access data volume of the new video is lower than the target value and the access data volume of the new video is continuously increasing, it can be determined that the new video requires traffic support.

[0030] It should be noted that when the candidate supported video is a new video, the supported video content pool can be referred to as a new film content pool.

[0031] Among them, the number of videos in the supported video content pool is not limited here. Optionally, the number of videos in the supported video content pool can be 20 to 30.

[0032] It should be noted that in an optional application scenario, the playback durations of the above multiple candidate supported videos can all be greater than the preset playback duration, so the above multiple candidate supported videos can all be referred to as long videos.

[0033] Among them, the specific content of the distribution efficiency index information is not limited here. Optionally, the distribution efficiency index information includes at least one of the following: click-through rate, average display playback duration, and effective playback rate.

[0034] Among them, the click-through rate can be understood as the percentage of the number of times a candidate promoted video is clicked and viewed within a preset historical time period to the total number of times the candidate promoted video is displayed. Or the click-through rate can also be referred to as (Click Through Rate, CTR), that is, CTR = number of video clicks / number of video displays × 100%; the average playtime per impression can be referred to as (Playtime Per Impression, PPI), and the average playtime per impression can be understood as: within a preset historical time period, the ratio of the total playtime of the candidate promoted video to the total number of times the candidate promoted video is displayed; the effective click-through rate can be referred to as (effective CTR, eCTR), and the effective click-through rate can be understood as the ratio of the number of times the candidate promoted video is continuously viewed for a duration exceeding the target duration to the total number of times the candidate promoted video is displayed within a preset historical time period.

[0035] In this way, the diversity and flexibility of the distribution efficiency index information can be increased, and the more index values included in the distribution efficiency index information, the more accurately the distribution efficiency of the candidate promoted video can be represented, that is, the accuracy of the distribution efficiency of the candidate promoted video is improved.

[0036] It should be noted that the preset historical time period is not specifically limited here. Optionally, the preset historical time period may include the time period before the current time window, and the specific length of the preset historical time period is not limited here. For example, the preset historical time period may be 1 hour, 4 hours, 12 hours, or 24 hours before the current time window.

[0037] Step 102: Screen at least one video to be promoted from the multiple candidate promoted videos based on the distribution efficiency index information.

[0038] It should be noted that the specific method of screening at least one video to be promoted from the multiple candidate promoted videos based on the distribution efficiency index information is not limited here.

[0039] As an optional implementation manner, the distribution efficiency index information is a distribution efficiency index value. The screening of at least one video to be promoted from the multiple candidate promoted videos based on the distribution efficiency index information includes:

[0040] Screen at least one candidate promoted video from the multiple candidate promoted videos whose distribution efficiency index information is greater than a preset threshold;

[0041] Determine the at least one candidate promoted video as the at least one video to be promoted.

[0042] Among them, the distribution efficiency index information is the distribution efficiency index value, which can be understood as: the distribution efficiency index information is directly represented in the form of the distribution efficiency index value. For example, the distribution efficiency index information can be the eCTR value.

[0043] Optionally, the distribution efficiency index information can also be text information. The index value can be extracted from the text information, and the extracted index value is determined as the distribution efficiency index value. When there are multiple index values extracted from the text information, the multiple index values can be weighted and calculated to obtain the distribution efficiency index value.

[0044] In the implementation manner of this application, by determining whether the distribution efficiency index information of each candidate supported video is greater than the preset threshold, and if the distribution efficiency index information of a certain candidate supported video is greater than the preset threshold, it means that the candidate supported video has been watched by the user more times. Therefore, the possibility that the candidate supported video is liked by the user is relatively high. In this way, at least one candidate supported video whose distribution efficiency index information is greater than the preset threshold is determined as at least one supported video, so that the determined supported video has a relatively high possibility of being liked and continuously watched by the user.

[0045] It should be noted that when the distribution efficiency index information only includes one index value, it can be directly calculated whether the index value of each candidate supported video is greater than the preset threshold, and the candidate supported video whose index value is greater than the preset threshold is determined as at least one supported video; optionally, when the distribution efficiency index information includes at least two index values, it can be determined respectively whether each index value is greater than the preset threshold, and at least one candidate supported video whose each index value is greater than the preset threshold is screened out from the multiple candidate supported videos, and at least one candidate supported video is determined as at least one supported video.

[0046] It should be noted that the specific determination method of the above preset threshold is not limited here. The preset threshold can be dynamically adjusted based on the distribution efficiency index information. For example, when the distribution efficiency index information in the first time period is less than the distribution efficiency index information in the second time period, and both the first time period and the second time period are time periods before the current time window, and the second time period is between the first time period and the current time window, that is, from the first time period through the second time period to the current time window, the distribution efficiency index information gradually increases, then the value of the preset threshold can be dynamically increased.

[0047] As an optional implementation manner, before screening out at least one candidate supported video whose distribution efficiency index information is greater than the preset threshold from the multiple candidate supported videos, the method further includes:

[0048] Sort the multiple candidate support videos in a preset order, where the preset order is from largest to smallest in terms of distribution efficiency index information, or the preset order is from smallest to largest in terms of distribution efficiency index information;

[0049] Determine a target support video at a target position from the multiple candidate support videos sorted in the preset order as the target support video, where the target position is a position other than the first position and the last position in the preset order;

[0050] Determine the distribution efficiency index information corresponding to the target support video as the preset threshold.

[0051] Among them, the target position is a position other than the first position and the last position in the preset order, that is, the target position can be a position between the first position and the last position in the preset order.

[0052] Optionally, the target position can be the median position in the preset order. Alternatively, the target position can be a specified position number in the preset order. For example, the target position can be the 30th position in the preset order.

[0053] In the embodiment of the present application, by sorting multiple candidate support videos in a preset order and determining the distribution efficiency index information corresponding to the target support video at the target position among the multiple candidate support videos sorted in the preset order as the preset threshold, in this way, the correlation between the preset threshold and the multiple candidate support videos can be improved, so that the accuracy of at least one video to be supported determined according to the preset threshold is higher.

[0054] It should be noted that the number of target support videos at the target position can be one or multiple. As an optional implementation manner, when the number of target support videos at the target position is multiple, when determining the preset threshold, one distribution efficiency index information can be randomly selected from the multiple distribution efficiency index information to be determined as the preset threshold, and the above multiple distribution efficiency index information can correspond to the multiple target support videos one by one.

[0055] As another optional implementation manner, the number of the target support videos is multiple, and the determining the distribution efficiency index information corresponding to the target support video as the preset threshold includes:

[0056] Calculate the average value of multiple distribution efficiency index information, and the multiple distribution efficiency index information corresponds to the multiple target support videos one by one;

[0057] Determine the average value as the preset threshold.

[0058] In the implementation manner of the present application, by calculating the average value of multiple distribution efficiency indicator information and determining the average value as the preset threshold, the accuracy of the determined preset threshold can be higher.

[0059] Step 103: determine the supported traffic of the video to be supported based on the distribution efficiency index information of the video to be supported and the total traffic to be supported obtained in advance, wherein the total traffic to be supported is the sum of the supported traffic of at least one video to be supported, and the supported traffic is the traffic allocated to the video to be supported when the video to be supported is promoted in the current time window.

[0060] It should be noted that the above steps 101 to 103 can be executed multiple times in a loop, and in the process of executing multiple times in a loop, due to the continuous change of time, the distribution efficiency index information of the video to be supported can be constantly changing with time, so the video to be supported screened according to the distribution efficiency index information can be changed in real time, which in turn causes the support flow of the determined video to be supported to also change in real time. That is to say, the embodiment of the present application provides a dynamic competition flow mechanism for the video to be supported, and the distribution efficiency index information of the video to be supported with good support effect will gradually become higher, and the support flow of the video to be supported determined according to the distribution efficiency index information will also increase accordingly, that is, the video to be supported can subsequently obtain more support flow; at the same time, since the video to be supported is updated in real time, the distribution efficiency index information of the video to be supported is also updated in real time. In this way, the method for determining the support flow of the video to be supported can have the characteristics of high sensitivity and high real-time performance.

[0061] The specific method of determining the supported flow of the video to be supported based on the distribution efficiency index information of the video to be supported and the pre-acquired total flow to be supported is not limited herein.

[0062] It should be noted that the traffic in the embodiment of the present application can be understood as data traffic, which refers to the amount of data transmission generated when other electronic devices access the electronic device in the embodiment of the present application. The traffic in the embodiment of the present application may include the above-mentioned supported traffic and the total traffic to be supported, etc.

[0063] Optionally, the total traffic to be supported is the sum of the supported traffic of at least one video to be supported, and the total traffic to be supported can also be referred to as the total supported traffic. The electronic device in the embodiment of the present application can be a server, and the total traffic to be supported can be part of the total traffic of other electronic devices accessing the server in the embodiment of the present application. For example, the total traffic of other electronic devices accessing the server in the embodiment of the present application can be expressed as S total The total flow to be supported can be expressed as S, S = β*S total, where β can be a constant, or β can also be related to the number of at least one video to be promoted, that is, the more the number of at least one video to be promoted, the more β is. And the at least one video to be promoted can be represented by item1, item2,..., item N is represented as

[0064] Optionally, the ratio of the distribution efficiency index information of the video to be promoted and the total distribution efficiency index information of at least one video to be promoted can be determined, and then the product of the total traffic of at least one video to be promoted and the ratio is determined as the promotion traffic of the video to be promoted.

[0065] As an optional implementation manner, determining the promotion traffic of the video to be promoted based on the distribution efficiency index information of the video to be promoted and the pre-obtained total promotion traffic includes:

[0066] Calculating the ratio of the distribution efficiency index information of the video to be promoted to the target value to obtain a target ratio; the target value is determined according to the sum of the distribution efficiency index information of the at least one video to be promoted;

[0067] Calculating the product of the target ratio and the total promotion traffic as the promotion traffic of the video to be promoted.

[0068] Among them, the sum of the distribution efficiency index information of at least one video to be promoted can be referred to as the total distribution efficiency index information of at least one video to be promoted.

[0069] For example, when the distribution efficiency index information of the video to be promoted is the eCTR value, the eCTR values of at least one video to be promoted item1, item2,..., item N can be represented by eCTR1, eCTR2,…, eCTR N respectively, and the eCTR value of a certain video to be promoted can be represented by eCTR i respectively.

[0070] Optionally, the step of calculating the ratio of the distribution efficiency index information of the video to be promoted to the target value to obtain the target ratio can be referred to the following description:

[0071] Normalize the eCTR values of at least one video to be promoted, and use the Softmax function to calculate the target ratio of each video to be promoted:

[0072]

[0073] Among them, H represents the set of at least one video to be promoted, that is, the number of at least one video to be promoted is H, and p iDenote the target ratio, where both i and j are used to represent the numbers of the videos to be supported.

[0074] Among them, calculating the product of the target ratio and the total traffic to be supported as the supported traffic of the video to be supported can be understood as: the supported traffic Si of the video to be supported i i = S·p i .

[0075] In the implementation manner of this application, calculating the product of the target ratio and the total traffic to be supported as the supported traffic of the video to be supported can improve the determination rate of the supported traffic of the video to be supported and the accuracy of the determined supported traffic of the video to be supported.

[0076] It should be noted that the specific determination method of the target value is not limited here. On the basis of the above method, as an optional implementation manner, calculating the ratio of the distribution efficiency index information of the video to be supported to the target value to obtain the target ratio includes:

[0077] Calculating the first product of the distribution efficiency index information of the video to be supported and the adjustment parameter, and determining the first product as the corrected distribution efficiency index information, where the adjustment parameter is used to represent the inclination degree of the supported traffic allocation of the video to be supported;

[0078] Calculating the second product of the distribution efficiency index information of each video to be supported among the at least one video to be supported and the adjustment parameter, and calculating the sum of at least one of the second products;

[0079] Determining the sum of at least one of the second products as the target value;

[0080] Calculating the ratio of the corrected distribution efficiency index information to the target value to obtain the target ratio.

[0081] Among them, the above implementation manner can refer to the following formula:

[0082]

[0083] Among them, α is used to represent the adjustment parameter, and α·eCTR i is used to represent the first product, and α·eCTR j is used to represent the second product, and ∑ j∈H exp(α·eCTR j ) is used to represent the sum of at least one of the second products.

[0084] It should be noted that the adjustment parameter is used to represent the inclination degree of the support traffic distribution of the video to be supported. That is, the larger the adjustment parameter, the larger the target ratio. Therefore, the product of the target ratio and the total support traffic is larger, that is, the support traffic of the video to be supported is larger. That is to say, when the adjustment parameter is larger, more support traffic is distributed to the video to be supported, that is, the inclination degree of the support traffic distribution of the video to be supported is larger.

[0085] In the embodiment of the present application, since the adjustment parameter is added and the adjustment parameter is used to represent the inclination degree of the support traffic distribution of the video to be supported, the determination method of the target value can be made more flexible.

[0086] It should be noted that since the video to be supported can correspond to multiple preset historical time periods, and the distribution efficiency index information of the video to be supported in multiple preset historical time periods can be different, the distribution efficiency index information of the video to be supported in multiple preset historical time periods can be obtained, and according to the above embodiments, multiple support traffic of the video to be supported can be calculated, and then the weighted sum of the multiple support traffic can be calculated, and the weighted sum is used as the corrected support traffic of the video to be supported, so that the error of the support traffic of the video to be supported can be reduced.

[0087] In the embodiment of the present application, through steps 101 to 103, the distribution efficiency index information of each candidate supported video among multiple candidate supported videos is obtained. The distribution efficiency index information is an index value used to represent the distribution efficiency of the candidate supported video within a preset historical time period. The preset historical time period includes the time period before the current time window; at least one video to be supported is screened from the multiple candidate supported videos based on the distribution efficiency index information; the support traffic of the video to be supported is determined based on the distribution efficiency index information of the video to be supported and the pre-obtained total support traffic. The total support traffic is the sum of the support traffic of the at least one video to be supported. The support traffic is the traffic allocated to the video to be supported when promoting the video to be supported in the current time window. In this way, based on the distribution efficiency index information of the video to be supported and the pre-obtained total support traffic, the support traffic of the video to be supported can be determined flexibly, thereby improving the flexibility of the determination method of the support traffic of the video to be supported.

[0088] As an optional implementation manner, after step 103, determining the support traffic of the video to be supported based on the distribution efficiency index information of the video to be supported and the pre-obtained total support traffic, the method further includes:

[0089] Determine a target user based on the display content of the video to be promoted, the promotion traffic, and a pre-stored user preference information table, where the user preference information table includes preference scores of multiple users for display content, and the target user is at least one of the multiple users;

[0090] Push the video to be promoted to the target user.

[0091] Among them, the display content of the video to be promoted can refer to the classification of the display content. For example, the display content can be classified as military, entertainment, ancient costume, modern, etc.

[0092] Among them, optionally, by determining how many target users can use the promotion traffic, the number of target users can be determined, that is, the number of target users can be determined through the promotion traffic. And the preference score corresponding to the display content of the personal information can be understood as the score for different display contents. For example, when the promotion traffic can be used by one target user, and the preference score of user A for military display content can be 0.8, the preference score of user A for entertainment display content can be 0.2, the preference score of user B for military display content can be 0.1, and the preference score of user B for entertainment display content can be 0.7. Then it can be determined that among military and entertainment, user A is more interested in military display content, and user B is more interested in entertainment display content. Therefore, when the display content of the video to be promoted is military, the video to be promoted can be pushed to user A, and when the display content of the video to be promoted is entertainment, the video to be promoted can be pushed to user B.

[0093] In the embodiment of the present application, a target user is determined based on the display content of the video to be promoted, the promotion traffic, and a pre-stored user preference information table, so that the accuracy of the determined target user is relatively high, that is, the target user is the audience of the video to be promoted. When the video to be promoted is pushed to the target user, the accuracy of the push of the video to be promoted can be relatively high.

[0094] As another alternative embodiment, the determining a target user according to the display content of the video to be promoted, the promotion traffic, and a pre-stored user preference information table includes:

[0095] Determine a user preference information table corresponding to the display content according to the display content of the video to be promoted;

[0096] Determine a corresponding target preference score in the user preference information table according to the promotion traffic;

[0097] Determine the user corresponding to the preference score greater than or equal to the target preference score in the user preference information table as the target user.

[0098] Among them, the user preference information tables corresponding to different display contents are different. For example, when the display content is military-related, it can correspond to the first user preference information table, and the first user preference information table can include the preference scores of different users for the military-related display content; when the display content is entertainment-related, it can correspond to the second user preference information table, and the second user preference information table can include the preference scores of different users for the entertainment-related display content.

[0099] Among them, in addition to storing the preference scores of multiple users for the display content, the user preference information table can also store the personal information of the users, as well as information such as the number of users corresponding to each preference score. The personal information of the users can refer to information such as the names and identifiers of the users, and the identifier can be a number, etc. In this way, based on the various information stored in the above user preference information table, a relationship graph of user preference scores and user proportions as shown in Figure 2 can be constructed. Figure 2 In [the graph], the abscissa is the preference score of the user for the display content, and this preference score can also be referred to as the prediction score. Figure 2 In [the graph], the ordinate is the distribution proportion of users corresponding to different prediction scores. In this way, through Figure 2 the distribution proportion of users corresponding to different preference scores can be accurately reflected.

[0100] It should be noted that when determining the target preference score corresponding to the user preference information table according to the support traffic, the following description can be referred to: The target distribution proportion of the target user can be determined according to the support traffic, and then in Figure 2 the users with the target distribution proportion are selected in the order of preference scores from large to small. In this way, the preference score corresponding to the last user in the target distribution proportion is determined as the target preference score.

[0101] For example: As shown in Figure 2 , through the above method, the preference score x of the user can be determined to be 0.428. In this way, the users with preference scores to the right of 0.428 (that is, the users shown in the D area in Figure 2 ) can all be determined as target users, and the distribution proportion of the users with preference scores to the right of 0.428 can exactly be the above target distribution proportion, and the video S to be supported is pushed to the target users i . In this way, it can be ensured that the users with preference scores to the right of 0.428 can roughly consume the support traffic of the video S to be supported. i And because the preference scores of the target users are all relatively high, the interest of the target users in the video S to be supported i is relatively high, so that the target users usually continue to watch the video S to be supported i . In this way, it enables the above target users to quickly consume the support traffic allocated to the video S to be supported.i Support traffic.

[0102] The implementation manner of the present application provides a method for determining the target user according to the display content of the video to be supported, the support flow and the user preference information table, and determining the target user according to the display content of the video to be supported, the support flow and the user preference information table can make the method for determining the target user more flexible.

[0103] As an optional implementation manner, before determining the target user according to the display content of the video to be supported, the supported traffic and a pre-stored user preference information table, the method further includes:

[0104] Obtaining personal information of the user;

[0105] Inputting the user's personal information into a preset model for scoring, and outputting the user's preference score, wherein the preset model is a pre-trained model for predicting the user's preference score;

[0106] The preference score is stored in the user preference information table.

[0107] Among them, the user's personal information may include gender, interests and hobbies, viewing history, etc.

[0108] The preset model can be called the user's personalized preference model M for content, and M can predict the personalized preference score m of any user u for displayed content i. ui Optionally, for each video content i to be supported, the electronic device will predict and record the content of each visiting user u in real time. x m(u x i), forming preference scores of all users regarding display content i, and recording them in the above user preference information table. At the same time, the following can be constructed based on the user preference information table: Figure 2 The relationship between user preference scores and user proportions is shown in Figure 2 , the relationship between the preference scores of all users showing content i and the user ratio can be expressed as Figure 2 As shown by line segment C in .

[0109] It should be noted that, since the preference score is obtained by the preset model M according to the user's personal information, the above preference score can also be referred to as the prediction score of the preset model M.

[0110] In the implementation manner of the present application, users are scored using a preset model, so that the user's preference score is obtained more accurately, and the preference score is stored in a user preference information table, which makes it easier to determine the target user later and improves the efficiency of determining the target user.

[0111] To more fully illustrate the above embodiments, a specific embodiment will be used for specific illustration below. The specific steps can be referred to Figure 3 as shown in Figure 3 as follows: The steps include:

[0112] Step 301, determine the supported video content pool;

[0113] Among them, after determining the supported video content pool, multiple contents can be obtained from the supported video content pool, and the multiple contents can refer to the candidate supported videos in the above embodiments;

[0114] Step 302, each content can participate in the dynamic competition for traffic and determine the traffic of each content in the current time window;

[0115] Among them, the dynamic competition for traffic can be understood as the process of determining the to-be-supported video from multiple candidate supported videos in the above embodiments, and the traffic in the current time window can be understood as the supported traffic in the current time window. Step 302 can be regarded as the process of statistical calculation of the dynamic competition effect of the traffic of the current content;

[0116] Step 303, determine the traffic allocation result of each supported content at the current time;

[0117] Among them, the traffic allocation result of each supported content at the current time can be understood as: the allocation result of the supported traffic of each supported content in the current time window, and the supported content can be understood as the to-be-supported video;

[0118] Step 304, convert the traffic demand into the score threshold of this supported content;

[0119] Among them, converting the traffic demand into the score threshold of this content can be understood as determining the target preference score of this supported content according to the supported traffic of a certain supported content. That is, the score threshold of this supported content can be understood as the target preference score;

[0120] Step 305, determine whether it is higher than the supported content threshold. If so, execute Step 309;

[0121] Among them, the supported content threshold can also be understood as the target preference score. Determining whether it is higher than the supported content threshold can be understood as: determining whether the user's preference score is greater than the target preference score. If the user's preference score is greater than the target preference score, then this user is the target user, and execute Step 309, that is, recommend the above supported content to this user. The specific generation method of the user's preference score can be referred to Step 307;

[0122] Step 306, the user enters the recommendation system;

[0123] Among them, when a user enters the recommendation system, content recommendations for fostering the user are performed, that is, step 307 is executed, and / or content recommendations for non-fostering the user are performed, that is, step 308 is executed;

[0124] Step 307: Traverse the fostering content, and obtain the preference score of the user for each piece of fostering content through the personalized preference model M;

[0125] Among them, the personalized preference model M can be a model trained using the user's behavior sample data. Traverse the fostering content, and each piece of fostering content can be scored through the personalized preference model M, so as to obtain the preference score of the user for each piece of fostering content;

[0126] Step 308: Recommend content using traditional algorithms;

[0127] Among them, recommending content using traditional algorithms can be understood as recommending content to the user using traditional algorithms, and the content recommended to the user using traditional algorithms can be referred to as non-fostering content. The traditional algorithms are not limited here. Optionally, the traditional algorithms can be algorithms for increasing the recommendation weight of content.

[0128] Step 309: Foster this content, and determine the final position according to the preference score and the sorting of non-fostering content;

[0129] Among them, foster this content, and the display order of this content and non-fostering content can be determined according to the preference score, so as to determine the final display position of this content, and the final display position can also be referred to as the last display position or the final position;

[0130] Step 310: Fusion;

[0131] Among them, the fusion in step 310 can be understood as performing an operation of fusing and sorting the fostering content and non-fostering content, that is, step 310 can be understood as performing an operation of sorting the fostering content and non-fostering content together;

[0132] Step 311: Display the fostering content and non-fostering content;

[0133] Among them, displaying the fostering content and non-fostering content can be understood as displaying the fostering content and non-fostering content according to the sorting determined in step 310.

[0134] See Figure 4 , Figure 4 is the structural diagram of the fostering traffic determination device provided by the embodiment of the present application. As Figure 4 shown, the fostering traffic determination device 400 includes:

[0135] The first acquisition module 401 is configured to acquire the distribution efficiency index information of each candidate supported video among multiple candidate supported videos, where the distribution efficiency index information is an index value used to represent the distribution efficiency of the candidate supported video within a preset historical time period, and the preset historical time period includes the time period before the current time window;

[0136] The screening module 402 is configured to screen at least one supported video from the multiple candidate supported videos based on the distribution efficiency index information;

[0137] The first determination module 403 is configured to determine the support traffic of the supported video based on the distribution efficiency index information of the supported video and the pre-acquired total support traffic, where the total support traffic is the sum of the support traffic of the at least one supported video, and the support traffic is the traffic allocated to the supported video when promoting the supported video in the current time window.

[0138] As an alternative implementation manner, the distribution efficiency index information is a distribution efficiency index value, and the screening module 402 includes:

[0139] The screening sub-module is configured to screen at least one candidate supported video whose distribution efficiency index information is greater than a preset threshold from the multiple candidate supported videos;

[0140] The first determination sub-module is configured to determine the at least one candidate supported video as the at least one supported video.

[0141] As an alternative implementation manner, the support traffic determination device 400 further includes:

[0142] The sorting module is configured to sort the multiple candidate supported videos in a preset order, where the preset order is the order from largest to smallest of the distribution efficiency index information, or the preset order is the order from smallest to largest of the distribution efficiency index information;

[0143] The second determination module is configured to determine the candidate supported video at the target position from the multiple candidate supported videos sorted in the preset order as the target supported video, where the target position is a position other than the first position and the last position in the preset order;

[0144] The third determination module is configured to determine the distribution efficiency index information corresponding to the target supported video as the preset threshold.

[0145] As an alternative implementation manner, the number of the target supported videos is multiple, and the third determination module includes:

[0146] The first calculation sub-module is used to calculate the average value of multiple distribution efficiency index information, and the multiple distribution efficiency index information corresponds to multiple target supported videos one by one;

[0147] The second determination sub-module is used to determine the average value as the preset threshold.

[0148] As an optional implementation manner, the distribution efficiency index information includes at least one of the following: click-through rate, average display duration, and effective playback rate.

[0149] As an optional implementation manner, the first determination module 403 includes:

[0150] The second calculation sub-module is used to calculate the ratio of the distribution efficiency index information of the video to be supported to the target value to obtain a target ratio; the target value is determined according to the sum of the distribution efficiency index information of at least one video to be supported;

[0151] The third calculation sub-module is used to calculate the product of the target ratio and the total traffic to be supported as the supported traffic of the video to be supported.

[0152] As an optional implementation manner, the second calculation sub-module includes:

[0153] The first calculation unit is used to calculate the first product of the distribution efficiency index information of the video to be supported and the adjustment parameter, and determine the first product as the corrected distribution efficiency index information, and the adjustment parameter is used to represent the inclination degree of the supported traffic allocation of the video to be supported;

[0154] The second calculation unit is used to calculate the second product of the distribution efficiency index information of each video to be supported in at least one video to be supported and the adjustment parameter, and calculate the sum of at least one second product;

[0155] The determination unit is used to determine the sum of at least one second product as the target value;

[0156] The third calculation unit is used to calculate the ratio of the corrected distribution efficiency index information to the target value to obtain a target ratio.

[0157] As an optional implementation manner, the supported traffic determination device 400 further includes:

[0158] The fourth determination module is used to determine the target user according to the display content of the video to be supported, the supported traffic and the pre-stored user preference information table, and the user preference information table includes the preference scores of multiple users for the display content, and the target user is at least one of the multiple users;

[0159] A push module, configured to push the video to be supported to the target user.

[0160] As an optional implementation manner, the fourth determination module includes:

[0161] A third determination sub-module, configured to determine a user preference information table corresponding to the display content according to the display content of the video to be supported;

[0162] A fourth determination sub-module, configured to determine a target preference score corresponding to the user preference information table according to the support traffic;

[0163] A fifth determination sub-module, configured to determine a user corresponding to a preference score greater than or equal to the target preference score in the user preference information table as the target user.

[0164] As an optional implementation manner, the support traffic determination device 400 further includes:

[0165] A second acquisition module, configured to acquire personal information of the user;

[0166] A scoring module, configured to input the personal information of the user into a preset model for scoring, and output a preference score of the user. The preset model is a model pre-trained for predicting the preference score of the user;

[0167] A storage module, configured to store the preference score in the user preference information table.

[0168] The support traffic determination device 400 can implement each process in the method embodiment of the present application Figure 1 and achieve the same beneficial effects. To avoid repetition, it will not be elaborated here.

[0169] The embodiment of the present application further provides an electronic device. Please refer to Figure 5 , the electronic device may include a processor 501, a memory 502, and a program 5021 stored on the memory 502 and executable on the processor 501. When the program 5021 is executed by the processor 501, it can implement Figure 1 any step in the corresponding method embodiment and achieve the same beneficial effects, which will not be elaborated here.

[0170] Those of ordinary skill in the art can understand that all or part of the steps for implementing the method in the above embodiments can be completed by hardware related to program instructions, and the program can be stored in a readable medium. The embodiment of the present application further provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it can implement the above Figure 1Any step in the corresponding method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0171] The storage medium described above, such as a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0172] The embodiment of the present application further provides a computer program product, including computer instructions, and when the computer instructions are executed by a processor, the above Figure 1 Any step in the corresponding method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0173] The above is the preferred implementation manner of the embodiment of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle described in the present application, several improvements and refinements can still be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A method for determining boost traffic, characterized in that Including: Obtaining the distribution efficiency index information of each candidate promoted video among a plurality of candidate promoted videos, where the distribution efficiency index information is an index value used to represent the distribution efficiency of the candidate promoted video within a preset historical time period, and the preset historical time period includes the time period before the current time window; Screening at least one video to be promoted from the plurality of candidate promoted videos based on the distribution efficiency index information; Determining the promotion traffic of the video to be promoted based on the distribution efficiency index information of the video to be promoted and the pre-obtained total promotion traffic, where the total promotion traffic is the sum of the promotion traffic of the at least one video to be promoted, and the promotion traffic is the traffic allocated to the video to be promoted when promoting the video to be promoted within the current time window.

2. The method according to claim 1, wherein The distribution efficiency index information is a distribution efficiency index value, and screening at least one video to be promoted from the plurality of candidate promoted videos based on the distribution efficiency index information includes: Screening at least one candidate promoted video from the plurality of candidate promoted videos whose distribution efficiency index information is greater than a preset threshold; Determining the at least one candidate promoted video as the at least one video to be promoted.

3. The method according to claim 2, wherein Before screening at least one candidate promoted video from the plurality of candidate promoted videos whose distribution efficiency index information is greater than a preset threshold, the method further includes: Sorting the plurality of candidate promoted videos in a preset order, where the preset order is the order from large to small of the distribution efficiency index information, or the preset order is the order from small to large of the distribution efficiency index information; Determining the candidate promoted video at the target position from the plurality of candidate promoted videos sorted in the preset order as the target promoted video, where the target position is the position other than the first position and the last position in the preset order; Determining the distribution efficiency index information corresponding to the target promoted video as the preset threshold.

4. The method according to claim 3, wherein The number of the target promoted videos is multiple, and determining the distribution efficiency index information corresponding to the target promoted video as the preset threshold includes: Calculating the average value of multiple distribution efficiency index information, where the multiple distribution efficiency index information corresponds to the multiple target promoted videos one by one; Determining the average value as the preset threshold.

5. The method according to claim 1, characterized in that, The distribution efficiency index information includes at least one of the following: click-through rate, average display play duration, and effective play rate.

6. The method according to claim 1, wherein Determining the promotion traffic of the video to be promoted based on the distribution efficiency index information of the video to be promoted and the pre-obtained total promotion traffic includes: Calculating the ratio of the distribution efficiency index information of the video to be promoted to the target value to obtain a target ratio; the target value is determined according to the sum of the distribution efficiency index information of the at least one video to be promoted; Calculating the product of the target ratio and the total promotion traffic as the promotion traffic of the video to be promoted.

7. The method according to claim 6, wherein Calculating the ratio of the distribution efficiency index information of the video to be promoted to the target value to obtain a target ratio includes: Calculate the first product of the distribution efficiency index information of the video to be supported and the adjustment parameter, and determine the first product as the corrected distribution efficiency index information, where the adjustment parameter is used to represent the inclination degree of the supported traffic allocation of the video to be supported; Calculate the second product of the distribution efficiency index information of each video to be supported among the at least one video to be supported and the adjustment parameter, and calculate the sum of at least one of the second products; Determine the sum of at least one of the second products as the target value; Calculate the ratio of the corrected distribution efficiency index information to the target value to obtain the target ratio.

8. The method according to any one of claims 1 to 7, characterized in that, After determining the supported traffic of the video to be supported based on the distribution efficiency index information of the video to be supported and the pre-obtained total supported traffic, the method further includes: Determine the target user according to the display content of the video to be supported, the supported traffic, and the pre-stored user preference information table, where the user preference information table includes the preference scores of multiple users for the display content, and the target user is at least one of the multiple users; Push the video to be supported to the target user.

9. The method according to claim 8, wherein The determining the target user according to the display content of the video to be supported, the supported traffic, and the pre-stored user preference information table includes: Determine the user preference information table corresponding to the display content according to the display content of the video to be supported; Determine the corresponding target preference score in the user preference information table according to the supported traffic; Determine the user corresponding to the preference score greater than or equal to the target preference score in the user preference information table as the target user.

10. The method according to claim 8, wherein Before determining the target user according to the display content of the video to be supported, the supported traffic, and the pre-stored user preference information table, the method further includes: Obtain the personal information of the user; Input the personal information of the user into a preset model for scoring, and output the preference score of the user, where the preset model is a model pre-trained for predicting the preference score of the user; Store the preference score into the user preference information table.

11. A fostering traffic determination device, characterized in that, Includes: The first acquisition module is used to acquire the distribution efficiency index information of each candidate supported video among multiple candidate supported videos, where the distribution efficiency index information is the index value used to represent the distribution efficiency of the candidate supported video within a preset historical time period, and the preset historical time period includes the time period before the current time window; The screening module is used to screen at least one video to be supported from the multiple candidate supported videos based on the distribution efficiency index information; The first determination module is used to determine the supported traffic of the video to be supported based on the distribution efficiency index information of the video to be supported and the pre-obtained total supported traffic, where the total supported traffic is the sum of the supported traffic of the at least one video to be supported, and the supported traffic is the traffic allocated to the video to be supported when promoting the video to be supported in the current time window.

12. An electronic device, comprising: A memory, a processor, and a program stored on the memory and executable on the processor; characterized in that the processor is configured to read the program in the memory to implement the steps in the supported traffic determination method according to any one of claims 1 to 10.

13. A readable storage medium for storing a program, characterized in that, When the program is executed by the processor, it implements the steps in the supported traffic determination method according to any one of claims 1 to 10.

14. A computer program product, characterized in that, It includes computer instructions, and when the computer instructions are executed by the processor, they implement the steps in the supported traffic determination method according to any one of claims 1 to 10.

Images