Content determination method and apparatus, electronic device, and storage medium

Abstract

This application provides a content determination method, apparatus, electronic device, and storage medium, belonging to the field of Internet technology. This application achieves automated updating of recommendation data by determining a preset number of target locations in a set of locations corresponding to a recommendation list, and determining multiple content pools corresponding to the recommendation list, wherein the content pools are used to provide content to the set of locations; for any target location among the preset number of target locations, determining the recommendation data and click data corresponding to the target location in the current period, updating the recommendation data based on the click data to obtain first recommendation data corresponding to the target location; and determining the content to be recommended to the target location in the next period from the multiple content pools based on the first recommendation data corresponding to the target location, thereby achieving automated updating of recommendation data without manual analysis and improving the efficiency of recommendation data updating.

Description

Technical Field

[0001] This application relates to the field of Internet technology, and in particular to a content determination method, apparatus, electronic device, and storage medium. Background Technology

[0002] Video is one of the main application areas of the Internet, with its coverage continuously increasing and its user base growing rapidly. Against this backdrop, how to efficiently push video content has become a hot research topic in the Internet field.

[0003] For example, regarding content delivery on television, video content is currently often presented in a list format, with multiple items included in the list, such as... Figure 1 The example push notification list includes six items, numbered 1-6. The specific video content pushed for each item is typically determined by data analysts based on historical data. However, manual analysis is inefficient and costly. Summary of the Invention

[0004] The purpose of this application is to provide a content determination method, apparatus, electronic device, and storage medium to solve the problem of low efficiency in manually determining push content. The specific technical solution is as follows:

[0005] Firstly, a content determination method is provided, the method comprising:

[0006] A preset number of target locations are determined in the location set corresponding to the recommendation list, and multiple content pools corresponding to the recommendation list are determined, wherein the content pools are used to provide content to the location set;

[0007] For any target location among the preset number of target locations, determine the recommended data and click data corresponding to the target location in the current period, update the recommended data based on the click data, and obtain the first recommended data corresponding to the target location;

[0008] Based on the first recommendation data corresponding to the target location, the content to be recommended to the target location in the next period is determined from multiple content pools.

[0009] Optionally, the recommendation data includes: recommendation probabilities corresponding to multiple content pools, and the click data includes: click-through rates corresponding to multiple content pools.

[0010] The step of updating the recommendation data based on the click data to obtain the first recommendation data corresponding to the target location includes:

[0011] Determine the target percentage of click-through rate for the multiple content pools;

[0012] Update the recommendation probabilities corresponding to the multiple content pools according to the target ratio to obtain the first probabilities corresponding to the multiple content pools, wherein the ratio of the first probabilities of the multiple content pools is consistent with the target ratio;

[0013] When multiple first probabilities meet preset conditions, the first probabilities corresponding to multiple content pools are determined as the first recommended data corresponding to the target location;

[0014] Specifically, when multiple first probabilities are all greater than or equal to a preset probability threshold, it is determined that multiple first probabilities meet the preset conditions; when there is a probability value among multiple first probabilities that is less than the preset probability threshold, it is determined that multiple first probabilities do not meet the preset conditions.

[0015] Optionally, the method further includes:

[0016] When multiple first probabilities do not meet the preset conditions, the difference between the first probability less than the preset probability threshold and the preset probability threshold is determined;

[0017] Based on the difference, multiple first probabilities are updated to obtain multiple second probabilities, and the multiple second probabilities are used as the first recommended data corresponding to the target location.

[0018] Optionally, the method further includes:

[0019] Based on a preset number of the first recommended data corresponding to the target location, determine the second recommended data for other locations in the recommendation list;

[0020] For each of the other locations, content to be recommended to the other location in the next period is determined from the multiple content pools based on the second recommendation data.

[0021] Optionally, determining the second recommendation data for other locations in the recommendation list based on a preset number of the first recommendation data corresponding to the target location includes:

[0022] For a preset number of target locations, the target location and the first recommended data corresponding to the target location are combined as a data combination, and the coordinate point corresponding to the data combination in a preset coordinate system is determined.

[0023] The multiple coordinate points are connected according to a preset connection method to obtain the target curve;

[0024] For other positions in the recommendation list, determine the second recommendation data corresponding to those other positions on the target curve.

[0025] Optionally, determining a preset number of target locations in the location set corresponding to the recommendation list includes:

[0026] Determine multiple locations in the location set and the exposure data corresponding to the multiple locations;

[0027] Based on the exposure data, the multiple locations are divided into multiple location intervals;

[0028] For the multiple location intervals, the positions located at both ends of the location intervals are determined as the target positions.

[0029] Optionally, dividing the multiple locations into multiple location intervals based on the exposure data includes:

[0030] For any one of the multiple locations, determine the first exposure data for that location and the second exposure data for the adjacent locations corresponding to that location;

[0031] Determine the difference between the first exposure data and the second exposure data;

[0032] When the difference is greater than a preset difference threshold, the position and the adjacent position are determined as the dividing line;

[0033] The multiple locations are divided into multiple location intervals according to the defined boundaries.

[0034] Secondly, a content determination apparatus is provided, the apparatus comprising:

[0035] The first determining module is used to determine a preset number of target locations in the location set corresponding to the recommendation list, and to determine multiple content pools corresponding to the recommendation list, wherein the content pools are used to provide content to the location set;

[0036] The second determining module is used to determine, for any target location among a preset number of target locations, the recommended data and click data corresponding to the target location in the current period, update the recommended data based on the click data, and obtain the first recommended data corresponding to the target location;

[0037] The third determining module is used to determine, based on the first recommendation data corresponding to the target location, the content to be recommended to the target location in the next period from multiple content pools.

[0038] Optionally, the recommendation data includes: recommendation probabilities corresponding to multiple content pools, and the click data includes: click-through rates corresponding to multiple content pools.

[0039] The second determining module is specifically used for:

[0040] Determine the target percentage of click-through rate for the multiple content pools;

[0041] Update the recommendation probabilities corresponding to the multiple content pools according to the target ratio to obtain the first probabilities corresponding to the multiple content pools, wherein the ratio of the first probabilities of the multiple content pools is consistent with the target ratio;

[0042] When multiple first probabilities meet preset conditions, the first probabilities corresponding to multiple content pools are determined as the first recommended data corresponding to the target location;

[0043] Specifically, when multiple first probabilities are all greater than or equal to a preset probability threshold, it is determined that multiple first probabilities meet the preset conditions; when there is a probability value among multiple first probabilities that is less than the preset probability threshold, it is determined that multiple first probabilities do not meet the preset conditions.

[0044] Optionally, the device further includes a fourth determining module, specifically used for:

[0045] When multiple first probabilities do not meet the preset conditions, the difference between the first probability less than the preset probability threshold and the preset probability threshold is determined;

[0046] Based on the difference, multiple first probabilities are updated to obtain multiple second probabilities, and the multiple second probabilities are used as the first recommended data corresponding to the target location.

[0047] Optionally, the device further includes a fifth determining module, specifically used for:

[0048] Based on a preset number of the first recommended data corresponding to the target location, determine the second recommended data for other locations in the recommendation list;

[0049] For each of the other locations, content to be recommended to the other location in the next period is determined from the multiple content pools based on the second recommendation data.

[0050] Optionally, the fifth determining module is further configured to:

[0051] For a preset number of target locations, the target location and the first recommended data corresponding to the target location are combined as a data combination, and the coordinate point corresponding to the data combination in a preset coordinate system is determined.

[0052] The multiple coordinate points are connected according to a preset connection method to obtain the target curve;

[0053] For other positions in the recommendation list, determine the second recommendation data corresponding to those other positions on the target curve.

[0054] Optionally, the first determining module is specifically used for:

[0055] Determine multiple locations in the location set and the exposure data corresponding to the multiple locations;

[0056] Based on the exposure data, the multiple locations are divided into multiple location intervals;

[0057] For the multiple location intervals, the positions located at both ends of the location intervals are determined as the target positions.

[0058] Optionally, the first determining module is further configured to:

[0059] For any one of the multiple locations, determine the first exposure data for that location and the second exposure data for the adjacent locations corresponding to that location;

[0060] Determine the difference between the first exposure data and the second exposure data;

[0061] When the difference is greater than a preset difference threshold, the position and the adjacent position are determined as the dividing line;

[0062] The multiple locations are divided into multiple location intervals according to the defined boundaries.

[0063] Thirdly, an electronic device is provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0064] Memory, used to store computer programs;

[0065] When a processor executes a program stored in memory, it implements any of the steps described in the first aspect.

[0066] Fourthly, a computer-readable storage medium is provided, characterized in that the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of any of the methods described in the first aspect.

[0067] Fifthly, a computer program product containing instructions is provided, which, when run on a computer, causes the computer to execute any of the content determination methods described above.

[0068] Beneficial effects of the embodiments in this application:

[0069] This application provides a content determination method, apparatus, electronic device, and storage medium. In this application, for a target position in the recommendation list, the recommendation data for the next period can be determined by using the recommendation data and click data corresponding to the target position in the current period. Then, in the next period, the recommendation data is used to determine the content to be recommended to the target position from multiple content pools, thereby realizing the automatic updating of recommendation data without the need for manual analysis and improving the efficiency of recommendation data updating.

[0070] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0071] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0072] Figure 1 A schematic diagram of the structure of a recommendation list provided in an embodiment of this application;

[0073] Figure 2 A flowchart illustrating a content determination method provided in an embodiment of this application;

[0074] Figure 3 A flowchart illustrating a content determination method provided in another embodiment of this application;

[0075] Figure 4 This application provides a graph obtained by connecting coordinate points corresponding to the content pool in an embodiment of the present application.

[0076] Figure 5 A flowchart illustrating a content determination method provided in another embodiment of this application;

[0077] Figure 6 This is a schematic diagram of the structure of a content determination device provided in an embodiment of this application;

[0078] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0079] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0080] The following will describe in detail a content determination method provided in the embodiments of this application, with reference to specific implementation methods. Figure 2 As shown, the specific steps are as follows:

[0081] S201, determine a preset number of target locations in the location set corresponding to the recommendation list, and determine multiple content pools corresponding to the recommendation list, wherein the content pools are used to provide content to the location set.

[0082] This application provides a content determination method that can be applied to determine content to be recommended to a target position in a recommendation list from multiple content pools. The recommendation list includes multiple positions for displaying recommended content, and these multiple positions form a position set corresponding to the recommendation list. The content pools are used to provide content to each position in the position set; that is, the content displayed at each position comes from multiple content pools corresponding to the recommendation list.

[0083] The content can be a category of media resources (such as documentaries, movies, or TV series) or a specific media resource (such as movie A). All content can be pre-divided into different content pools according to different calculation methods, such as by number of views, by the age of the film, and by popular tags.

[0084] For example, assuming the maximum number of contents stored in the content pool is 200, and taking videos as a category, the number of videos in that content is i.

[0085] When calculating based on play count: Retrieve the historical starting count for each video: `video_view`. Then, the corresponding play count for that content is:

[0086]

[0087] All content is sorted in descending order of view count, and the top 200 pieces are assigned to the most popular video content pool.

[0088] When calculating the age of a video based on its online time: Obtain the online time of each video: online_time. Use the time elapsed since the last video as the measure of its age: current_time - online_time. Then, the age of the content is calculated as follows:

[0089]

[0090] All content is sorted in descending order of play count, and the top 200 pieces are allocated to the latest video content pool.

[0091] When calculating scores based on popular tags: Retrieve all currently popular tags across the entire network and match them with the video tags of each video within the content. Each match between a video tag and a popular tag accumulates 1 point. This process is used to calculate the score for all content. All content is then sorted in descending order of score, and the top 200 pieces are grouped into a popular tag video content pool.

[0092] Specifically, the correspondence between the recommendation list and multiple content pools, as well as the correspondence between the number of positions and a preset number, are pre-set and stored. When a recommendation list is obtained, the multiple content pools corresponding to the recommendation list are determined, and the corresponding preset number is determined based on the number of positions in the set of positions corresponding to the recommendation list.

[0093] Furthermore, the target positions are determined according to a preset number, where the positions at the beginning and end of the recommended list are the target positions by default.

[0094] S202, for any target location among the preset number of target locations, determine the recommended data and click data corresponding to the target location in the current period, update the recommended data based on the click data, and obtain the first recommended data corresponding to the target location.

[0095] S203, based on the first recommendation data corresponding to the target location, determine the content to be recommended to the target location in the next period from multiple content pools.

[0096] In practical applications, each position in the recommendation list is configured with recommendation data in each recommendation cycle. This recommendation data serves as the basis for determining the content to be pushed to that position from multiple content pools in the current cycle; the click data represents the actual click situation corresponding to that position in the current cycle.

[0097] The recommendation data may include: the probability of each content pool pushing content to the target location, wherein the sum of the recommendation probabilities of multiple content pools pushing content to the target location is 1; the click data may include: the click-through rate corresponding to the display of content pushed by each content pool at the target location.

[0098] For each target location, recommended data can be determined based on pre-configured data, and click data for that target location can be collected at the end of the current period. Then, the recommended data is updated based on the click data to obtain the first recommended data for the target location.

[0099] Then, based on the recommendation probability of each content pool in the first recommendation data to recommend content to the target location, the target content pool for recommending content to the target location is determined. Content in the target content pool that is not displayed in the recommendation list is determined as candidate content. The candidate content is sorted according to the sorting method corresponding to the content pool (such as in descending order of play count). The candidate content ranked first is determined as the content recommended to the target location in the next cycle.

[0100] Furthermore, after pushing the corresponding content to each location, the push status of that content is recorded, and the system determines whether the content has been displayed in the recommendation list based on this record. For example, the index function `index` can be used to mark the location where the content was pushed.

[0101] For example, the recommendation list corresponds to three content pools A, B, and C. The recommendation data corresponding to position d in the recommendation list is A=20%, B=30%, and C=50%. That is, the recommendation probability of content pool A pushing content to position d is 20%, the recommendation probability of content pool B pushing content to position d is 30%, and the recommendation probability of content pool C pushing content to position d is 50%.

[0102] When recommending content to position d, the integers 1-100 can be divided into three intervals: 1-20, 21-50, and 51-100, according to the proportions of 20%, 30%, and 50%. The random function Random(100) is used to generate the corresponding random integers within 100. Based on the interval in which the integer result falls, the target content pool is determined, and the content recommended to position d in the next cycle is selected from the target content pool.

[0103] In this embodiment of the application, for a target position in the recommendation list, the recommendation data for the next period can be determined by using the recommendation data and click data corresponding to the target position in the current period. Then, in the next period, the content to be recommended to the target position can be determined from multiple content pools using the recommendation data, thereby realizing the automatic updating of recommendation data without the need for manual analysis and improving the efficiency of recommendation data updating.

[0104] In yet another embodiment of this application, as Figure 3 As shown, step S202 may include the following steps:

[0105] S301, determine the target percentage of click-through rate for the multiple content pools.

[0106] S302, update the recommendation probabilities corresponding to the multiple content pools according to the target ratio to obtain the first probabilities corresponding to the multiple content pools, wherein the ratio of the first probabilities of the multiple content pools is consistent with the target ratio.

[0107] S303, when multiple first probabilities meet preset conditions, determine the first probabilities corresponding to the multiple content pools as the first recommended data corresponding to the target location.

[0108] Specifically, when multiple first probabilities are all greater than or equal to a preset probability threshold, it is determined that multiple first probabilities meet the preset conditions; when there is a probability value among multiple first probabilities that is less than the preset probability threshold, it is determined that multiple first probabilities do not meet the preset conditions.

[0109] In this embodiment of the application, a target proportion of the click-through rate of multiple content pools is determined, and the recommendation probability corresponding to the multiple content pools is updated according to the target proportion to obtain the first probability corresponding to the multiple content pools, wherein the proportion of the first probability corresponding to the multiple content pools is consistent with the target proportion.

[0110] For example, when ten devices recommend content through this recommendation list, the recommendation probabilities for the three content pools A, B, and C are A = 20%, B = 30%, and C = 50%. Ideally, two devices (1-2) will display content recommended by content pool A at their target positions, three devices (3-5) will display content recommended by content pool B at their target positions, and five devices (5-10) will display content recommended by content pool C at their target positions.

[0111] During actual user viewing, the click-through rate (CTR) of the target position on device (1-2) was 5%, the CTR of the target position on device (3-5) was 35%, and the CTR of the target position on device (5-10) was 60%. That is, the CTRs of the three content pools A, B, and C were 5%, 35%, and 60%, respectively. The target ratio was 5%:35%:60%. The recommendation probabilities of the three content pools A, B, and C were updated according to this target ratio to obtain the first probabilities of the three content pools A, B, and C, which were 5%, 35%, and 60%, respectively.

[0112] Then, it is determined whether multiple first probabilities meet preset conditions. When multiple first probabilities are all greater than or equal to a preset probability threshold, it is determined that multiple first probabilities meet the preset conditions. When there is a probability value among multiple first probabilities that is less than the preset probability threshold, it is determined that multiple first probabilities do not meet the preset conditions.

[0113] When multiple first probabilities meet preset conditions, the first probabilities corresponding to multiple content pools are determined as the first recommended data corresponding to the target location.

[0114] In this embodiment, the click-through rates of multiple content pools can be used to update the recommendation probabilities of multiple content pools. The update process does not require human intervention, which improves processing efficiency.

[0115] In yet another embodiment of this application, step S202 may further include the following steps:

[0116] Step 1: When multiple first probabilities do not meet the preset conditions, determine the difference between the first probability less than the preset probability threshold and the preset probability threshold.

[0117] Step 2: Based on the difference, update multiple first probabilities to obtain multiple second probabilities, and use the multiple second probabilities as the first recommended data corresponding to the target position.

[0118] In this embodiment of the application, when multiple first probabilities do not meet the preset conditions, the difference between the first probability less than the preset probability threshold and the preset probability threshold is determined. For the first probability less than the preset probability threshold, the first probability less than the preset probability threshold is updated to the same value as the preset probability threshold based on the difference, and the corresponding second probability is obtained.

[0119] For a first probability that is greater than or equal to a preset probability threshold, if there is only one such probability, the second probability is obtained by subtracting the difference from the first probability. For example, if Ai% = 5%, Bi% = 5%, and Ci% = 90% before the update, and the preset probability threshold is 10%, then Ai% = 10%, Bi% = 10%, and Ci% = 80% after the update.

[0120] If there are multiple first probabilities, the first probabilities that are greater than or equal to the preset probability threshold are reduced proportionally until the sum of the reduction values ​​equals the difference, thus obtaining the corresponding second probabilities. For example, before the update, Ai% = 5%, Bi% = 30%, Ci% = 65%, and the preset probability threshold is 10%, then after the update, Ai% = 10%, Bi% = 28.4%, and Ci% = 61.6%.

[0121] In this embodiment, the click-through rates of multiple content pools can be used to update the recommendation probabilities of multiple content pools. The update process does not require human intervention, which improves processing efficiency.

[0122] In yet another embodiment of this application, the method may further include the following steps:

[0123] Step 1: Based on the first recommended data corresponding to the target location with a preset number of locations, determine the second recommended data for other locations in the recommended list.

[0124] Step two: For each of the other locations, determine the content to be recommended to the other location in the next period based on the second recommendation data from multiple content pools.

[0125] In this embodiment of the application, for positions other than the target position in the recommendation list, the second recommendation data corresponding to the other positions can be determined using the first recommendation data corresponding to the target position.

[0126] As one implementation method, second recommendation data for other locations can be determined in the following way:

[0127] For a preset number of target locations, the target location and the first recommended data corresponding to the target location are combined as a data combination, and the coordinate point corresponding to the data combination in a preset coordinate system is determined; multiple coordinate points are connected according to a preset connection method to obtain a target curve; for other locations in the recommendation list, the second recommended data corresponding to the other locations is determined on the target curve.

[0128] In this embodiment, for each target location and its corresponding first recommended data, the target location can be used as the abscissa of a preset coordinate system, and its corresponding first recommended data can be used as the ordinate of the preset coordinate system to determine its corresponding coordinate point. Multiple coordinate points can be connected by using the method of obtaining Bézier curves to connect multiple coordinate points, or by using smooth curves to connect multiple coordinate points to obtain a target curve. For each other location, the location is used as the abscissa, and the corresponding ordinate is found on the target curve to obtain its corresponding second recommended data.

[0129] When a target location corresponds to multiple content pools, the first recommendation data includes the recommendation probability of multiple content pools recommending content to the target location. With the target location as the horizontal axis and the probability corresponding to each content pool as the vertical axis, multiple coordinate points corresponding to each content pool can be obtained. Therefore, multiple curves corresponding to multiple content pools can be obtained. Similarly, the recommendation probability of multiple content pools corresponding to each location can be obtained from these curves.

[0130] For example, when the target location corresponds to three content pools: the hottest video pool, the newest video pool, and the popular tag video pool, the following can be obtained: Figure 4 The three curves shown represent the recommendation probability of each position in the popular video pool, the recommendation probability of each position in the hottest video pool, and the recommendation probability of each position in the newest video pool.

[0131] In this embodiment, second recommendation data for other locations in the recommendation list can be determined based on the first recommendation data corresponding to the target location. Then, based on the second recommendation data, content to be recommended to other locations in the next cycle can be determined from multiple content pools. This eliminates the need to calculate recommendation data for each location based on click data, reducing computational load and improving processing efficiency.

[0132] In yet another embodiment of this application, as Figure 5 As shown, step S201 may include the following steps:

[0133] S501, determine multiple locations in the location set and the exposure data corresponding to the multiple locations.

[0134] S502, based on the exposure data, the multiple locations are divided into multiple location intervals.

[0135] S503, for the multiple location intervals, the positions located at both ends of the location intervals are determined as the target positions.

[0136] In practical applications, multiple locations in the recommendation list are generally arranged in a top-to-bottom order, and their corresponding exposure data also show a decreasing trend from top to bottom. In this embodiment, multiple locations in the location set and their corresponding exposure data can be determined. Based on the exposure data, the multiple locations are divided into multiple location intervals. For each location interval, the locations at both ends of the interval are determined as target locations.

[0137] As one implementation method, the exposure data interval can be pre-divided, and the position can be divided according to the interval to which the exposure data of each position belongs, resulting in multiple position intervals.

[0138] As another implementation, for any one of the multiple locations, a first exposure data for the location and a second exposure data for the adjacent location corresponding to the location are determined; the difference between the first exposure data and the second exposure data is determined; when the difference is greater than a preset difference threshold, the location and the adjacent location are determined as a dividing boundary; the multiple locations are divided into multiple location intervals according to the dividing boundary.

[0139] In this implementation, when the exposure data shows significant changes, multiple locations are divided according to the nodes of significant changes.

[0140] It should be noted that each location interval includes multiple consecutive locations. If the exposure data corresponding to a certain location does not conform to the decreasing trend, it is regarded as error data and its impact on the division rules is not considered when dividing the location interval.

[0141] In this embodiment, the location range can be automatically divided based on the exposure data to determine the target location, eliminating the need for manual determination and improving efficiency.

[0142] Based on the same technical concept, embodiments of this application also provide a content determination device, such as... Figure 6 As shown, the device includes:

[0143] The first determining module 601 is used to determine a preset number of target locations in the location set corresponding to the recommendation list, and to determine multiple content pools corresponding to the recommendation list, wherein the content pools are used to provide content to the location set;

[0144] The second determining module 602 is used to determine, for any target location among a preset number of target locations, the recommended data and click data corresponding to the target location in the current period, update the recommended data based on the click data, and obtain the first recommended data corresponding to the target location;

[0145] The third determining module 603 is used to determine, based on the first recommendation data corresponding to the target location, the content to be recommended to the target location in the next cycle from multiple content pools.

[0146] Optionally, the recommendation data includes: recommendation probabilities corresponding to multiple content pools, and the click data includes: click-through rates corresponding to multiple content pools.

[0147] The second determining module is specifically used for:

[0148] Determine the target percentage of click-through rate for the multiple content pools;

[0149] Update the recommendation probabilities corresponding to the multiple content pools according to the target ratio to obtain the first probabilities corresponding to the multiple content pools, wherein the ratio of the first probabilities of the multiple content pools is consistent with the target ratio;

[0150] When multiple first probabilities meet preset conditions, the first probabilities corresponding to multiple content pools are determined as the first recommended data corresponding to the target location;

[0151] Specifically, when multiple first probabilities are all greater than or equal to a preset probability threshold, it is determined that multiple first probabilities meet the preset conditions; when there is a probability value among multiple first probabilities that is less than the preset probability threshold, it is determined that multiple first probabilities do not meet the preset conditions.

[0152] Optionally, the device further includes a fourth determining module, specifically used for:

[0153] When multiple first probabilities do not meet the preset conditions, the difference between the first probability less than the preset probability threshold and the preset probability threshold is determined;

[0154] Based on the difference, multiple first probabilities are updated to obtain multiple second probabilities, and the multiple second probabilities are used as the first recommended data corresponding to the target location.

[0155] Optionally, the device further includes a fifth determining module, specifically used for:

[0156] Based on a preset number of the first recommended data corresponding to the target location, determine the second recommended data for other locations in the recommendation list;

[0157] For each of the other locations, content to be recommended to the other location in the next period is determined from the multiple content pools based on the second recommendation data.

[0158] Optionally, the fifth determining module is further configured to:

[0159] For a preset number of target locations, the target location and the first recommended data corresponding to the target location are combined as a data combination, and the coordinate point corresponding to the data combination in a preset coordinate system is determined.

[0160] The multiple coordinate points are connected according to a preset connection method to obtain the target curve;

[0161] For other positions in the recommendation list, determine the second recommendation data corresponding to those other positions on the target curve.

[0162] Optionally, the first determining module is specifically used for:

[0163] Determine multiple locations in the location set and the exposure data corresponding to the multiple locations;

[0164] Based on the exposure data, the multiple locations are divided into multiple location intervals;

[0165] For the multiple location intervals, the positions located at both ends of the location intervals are determined as the target positions.

[0166] Optionally, the first determining module is further configured to:

[0167] For any one of the multiple locations, determine the first exposure data for that location and the second exposure data for the adjacent locations corresponding to that location;

[0168] Determine the difference between the first exposure data and the second exposure data;

[0169] When the difference is greater than a preset difference threshold, the position and the adjacent position are determined as the dividing line;

[0170] The multiple locations are divided into multiple location intervals according to the defined boundaries.

[0171] In this embodiment of the application, for a target position in the recommendation list, the recommendation data for the next period can be determined by using the recommendation data and click data corresponding to the target position in the current period. Then, in the next period, the content to be recommended to the target position can be determined from multiple content pools using the recommendation data, thereby realizing the automatic updating of recommendation data without the need for manual analysis and improving the efficiency of recommendation data updating.

[0172] Based on the same technical concept, embodiments of this application also provide an electronic device, such as... Figure 7 As shown, it includes a processor 111, a communication interface 112, a memory 113, and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 communicate with each other through the communication bus 114.

[0173] Memory 113 is used to store computer programs;

[0174] When processor 111 executes a program stored in memory 113, it performs the following steps:

[0175] A preset number of target locations are determined in the location set corresponding to the recommendation list, and multiple content pools corresponding to the recommendation list are determined, wherein the content pools are used to provide content to the location set;

[0176] For any target location among the preset number of target locations, determine the recommended data and click data corresponding to the target location in the current period, update the recommended data based on the click data, and obtain the first recommended data corresponding to the target location;

[0177] Based on the first recommendation data corresponding to the target location, the content to be recommended to the target location in the next period is determined from multiple content pools.

[0178] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0179] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0180] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0181] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0182] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of any of the above-described content determination methods.

[0183] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform any of the content determination methods described in the above embodiments.

[0184] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0185] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0186] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A content determination method characterized by comprising: The method includes: A preset number of target locations are determined in the location set corresponding to the recommendation list, and multiple content pools corresponding to the recommendation list are determined, wherein the content pools are used to provide content to the location set; For any target location among the preset number of target locations, determine the recommended data and click data corresponding to the target location in the current period, update the recommended data based on the click data, and obtain the first recommended data corresponding to the target location; Based on the first recommendation data corresponding to the target location, determine the content to be recommended to the target location in the next period from multiple content pools; The recommendation data includes: recommendation probabilities corresponding to multiple content pools; the click data includes: click-through rates corresponding to multiple content pools. The step of updating the recommendation data based on the click data to obtain the first recommendation data corresponding to the target location includes: Determine the target percentage of click-through rate for the multiple content pools; Update the recommendation probabilities corresponding to the multiple content pools according to the target ratio to obtain the first probabilities corresponding to the multiple content pools, wherein the ratio of the first probabilities of the multiple content pools is consistent with the target ratio; When multiple first probabilities meet preset conditions, the first probabilities corresponding to multiple content pools are determined as the first recommended data corresponding to the target location; Specifically, when multiple first probabilities are all greater than or equal to a preset probability threshold, it is determined that multiple first probabilities meet the preset conditions; when there is a probability value among multiple first probabilities that is less than the preset probability threshold, it is determined that multiple first probabilities do not meet the preset conditions.

2. The method of claim 1, wherein, The method further includes: When multiple first probabilities do not meet the preset conditions, the difference between the first probability less than the preset probability threshold and the preset probability threshold is determined; Based on the difference, multiple first probabilities are updated to obtain multiple second probabilities, and the multiple second probabilities are used as the first recommended data corresponding to the target location.

3. The method of claim 1, wherein, The method further includes: Based on a preset number of the first recommended data corresponding to the target location, determine the second recommended data for other locations in the recommendation list; For each of the other locations, content to be recommended to the other location in the next period is determined from the multiple content pools based on the second recommendation data.

4. The method of claim 3, wherein, The process of determining second recommended data for other locations in the recommended list based on a preset number of the first recommended data corresponding to the target locations includes: For a preset number of target locations, the target location and the first recommended data corresponding to the target location are combined as a data combination, and the coordinate point corresponding to the data combination in a preset coordinate system is determined. The multiple coordinate points are connected according to a preset connection method to obtain the target curve; For other positions in the recommendation list, determine the second recommendation data corresponding to those other positions on the target curve.

5. The method of claim 1, wherein, Determining a preset number of target locations from the location set corresponding to the recommendation list includes: Determine multiple locations in the location set and the exposure data corresponding to the multiple locations; Based on the exposure data, the multiple locations are divided into multiple location intervals; For the multiple location intervals, the positions located at both ends of the location intervals are determined as the target positions.

6. The method of claim 5, wherein, The step of dividing the multiple locations into multiple location intervals based on the exposure data includes: For any one of the multiple locations, determine the first exposure data for that location and the second exposure data for the adjacent locations corresponding to that location; Determine the difference between the first exposure data and the second exposure data; When the difference is greater than a preset difference threshold, the position and the adjacent position are determined as the dividing line; The multiple locations are divided into multiple location intervals according to the defined boundaries.

7. A content determination apparatus characterized by comprising: The device includes: The first determining module is used to determine a preset number of target locations in the location set corresponding to the recommendation list, and to determine multiple content pools corresponding to the recommendation list, wherein the content pools are used to provide content to the location set; The second determining module is used to determine, for any target location among a preset number of target locations, the recommended data and click data corresponding to the target location in the current period, update the recommended data based on the click data, and obtain the first recommended data corresponding to the target location; The third determining module is used to determine, based on the first recommendation data corresponding to the target location, the content to be recommended to the target location in the next period from multiple content pools; The recommendation data includes: recommendation probabilities corresponding to multiple content pools; the click data includes: click-through rates corresponding to multiple content pools. The step of updating the recommendation data based on the click data to obtain the first recommendation data corresponding to the target location includes: Determine the target percentage of click-through rate for the multiple content pools; Update the recommendation probabilities corresponding to the multiple content pools according to the target ratio to obtain the first probabilities corresponding to the multiple content pools, wherein the ratio of the first probabilities of the multiple content pools is consistent with the target ratio; When multiple first probabilities meet preset conditions, the first probabilities corresponding to multiple content pools are determined as the first recommended data corresponding to the target location; Specifically, when multiple first probabilities are all greater than or equal to a preset probability threshold, it is determined that multiple first probabilities meet the preset conditions; when there is a probability value among multiple first probabilities that is less than the preset probability threshold, it is determined that multiple first probabilities do not meet the preset conditions.

8. An electronic device, comprising: It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the steps of the method described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1-6.

Images