Video recommendation method and device, equipment, storage medium and program product

Abstract

The invention discloses a video recommendation method and device, equipment, a storage medium and a program product, and relates to the technical field of video processing, and the method comprises the steps: obtaining the feature data of each screen in a current display interface, and enabling the screens, obtained through screen splitting, of the current display interface to comprise a main screen and at least one auxiliary screen; sorting the main screen and the auxiliary screen according to the feature data; under the condition that the ranking of the main screen is the highest, the display mode of a current display interface is kept; and under the condition that the sequence of the secondary screens is the highest, recommending the secondary screen with the highest sequence to a user in a first mode. In this way, the most wonderful content can be presented to the user friendly.

Description

Technical Field

[0001] The present application relates to the field of video processing technology, and in particular to a video recommendation method, device, equipment, storage medium and program product. Background Art

[0002] In today's era of Internet video prevalence, multiple videos (such as games) are often played at the same time (such as live broadcasts), and users often face the situation of wanting to watch the broadcast (live broadcast) of multiple videos (games) at the same time. In related technologies, the multi-channel simultaneous viewing method can satisfy users' need to watch multiple videos (such as games) at the same time, which is specifically achieved through the main and sub-screen method. This method can meet the needs of users to watch multiple videos (such as games) at the same time, but because the screen of the sub-screen is relatively small, especially when there are multiple sub-screens, users may often miss the wonderful moments of the sub-screen program, and it is impossible to present the most exciting content at a certain moment in the multiple videos to users in a more friendly way. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a video recommendation method, apparatus, device, storage medium and program product, so as to solve the problem that the most exciting content at a certain moment in multiple videos cannot be presented to the user in a friendly manner.

[0004] In a first aspect, in order to achieve the above-mentioned purpose, an embodiment of the present application provides a video recommendation method, comprising:

[0005] Respectively acquiring characteristic data of each screen in the current display interface, wherein the screens obtained by splitting the current display interface include a main screen and at least one sub-screen;

[0006] sorting the main screen and the sub-screen according to the characteristic data;

[0007] When the main screen has the highest ranking, maintaining the display mode of the current display interface;

[0008] In the case where the slave screen has the highest ranking, the slave screen with the highest ranking is recommended to the user in a first manner.

[0009] Optionally, the sorting the main screen and the slave screen according to the characteristic data includes:

[0010] According to each of the feature data, obtaining a viewing growth rate score, a feature data score and a wonderfulness score of the screen corresponding to the feature data;

[0011] The main screen and the sub-screen are sorted according to the viewing growth rate score, the characteristic data score and the wonderfulness score.

[0012] Optionally, obtaining, according to the feature data, a viewing growth rate score, a feature data score, and an exciting degree score of a screen corresponding to the feature data, includes:

[0013] Calculating the viewing popularity growth rate of the screen corresponding to the feature data according to the viewing popularity information in the two adjacent feature data obtained;

[0014] Calculate the viewing growth rate score according to the viewing popularity growth rate and a pre-configured first weight;

[0015] Determining key moment feature data according to key moment information in the feature data, and determining star moment feature data according to star moment information in the feature data;

[0016] Determining, according to the video level in the feature data, a second weight corresponding to the key moment feature data and a third weight corresponding to the star moment feature data;

[0017] Calculating the feature data score according to the key moment feature data, the star moment feature data, the second weight, and the third weight;

[0018] Calculating the wonderfulness score according to the viewing growth rate score and the feature data score. Optionally, sorting the main screen and the secondary screen according to the viewing growth rate score, the feature data score and the wonderfulness score includes:

[0019] According to the wonderfulness score, the main screen and the secondary screen are sorted in descending order to obtain a first sorting result;

[0020] In the first sorting result, when the main screen has the highest sorting, determining the sorting result of the main screen and the slave screen as the first sorting result;

[0021] In the first sorting result, if the slave screen sorted before the main screen exists, acquiring a first slave screen set, the first slave screen set including the slave screen sorted before the main screen;

[0022] According to the viewing growth rate score and the feature data score, the respective slave screens in the first slave screen set and the main screen are sorted to obtain the sorting results of the main screen and the slave screens.

[0023] Optionally, sorting the slave screens in the first slave screen set and the main screen according to the viewing growth rate score and the feature data score to obtain the sorting results of the main screen and the slave screens includes:

[0024] Compare the viewing growth rate score and the feature data score of the main screen with those of each of the secondary screens in the first secondary screen set, respectively, to obtain a second secondary screen set, wherein the viewing growth rate score and the feature data score of each of the secondary screens in the second secondary screen set are both greater than the viewing growth rate score and the feature data score of the main screen;

[0025] In the case that the second secondary screen set is not an empty set, the secondary screens and the main screen in the second secondary screen set are sorted in descending order according to the wonderfulness score to obtain a second sorting result, wherein the second sorting result is a sorting result of the main screen and the secondary screen;

[0026] In the case that the second slave screen set is an empty set, the sorting result of the main screen and the slave screen is determined to be the first sorting result.

[0027] Optionally, recommending the slave screen with the highest ranking to the user in a first manner includes at least one of the following:

[0028] When the ranking result of the main screen and the slave screen is the first ranking result, the slave screen with the highest ranking is recommended to the user by highlighting and flashing the frame of the slave screen with the highest ranking;

[0029] When the ranking result of the main screen and the slave screen is the second ranking result, the slave screen with the highest ranking is recommended to the user by automatically switching the main screen and the slave screen with the highest ranking.

[0030] In a second aspect, in order to achieve the above-mentioned purpose, an embodiment of the present application provides a video recommendation device, including:

[0031] An acquisition module, used to respectively acquire characteristic data of each screen in the current display interface, wherein the screens obtained by splitting the current display interface include a main screen and at least one sub-screen;

[0032] A sorting module, used for sorting the main screen and the sub-screen according to the characteristic data;

[0033] A display module, used for maintaining the display mode of the current display interface when the main screen is ranked highest;

[0034] The recommendation module is used to recommend the slave screen with the highest ranking to the user in a first manner when the slave screen has the highest ranking.

[0035] In a third aspect, in order to achieve the above-mentioned purpose, an embodiment of the present application provides a video recommendation device, including: a transceiver, a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the video recommendation method described in the first aspect is implemented.

[0036] In a fourth aspect, in order to achieve the above-mentioned purpose, an embodiment of the present application provides a readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the video recommendation method as described in the first aspect is implemented.

[0037] In a fifth aspect, in order to achieve the above-mentioned purpose, an embodiment of the present application provides a computer program product, including computer instructions, which, when executed by a processor, implement the video recommendation method as described in the first aspect.

[0038] The above technical solution of the present application has at least the following beneficial effects:

[0039] In the video recommendation method of the embodiment of the present application, first, characteristic data of each screen in the current display interface is obtained respectively, wherein the screens obtained by splitting the current display interface include a main screen and at least one sub-screen; secondly, the main screen and the sub-screen are sorted according to the characteristic data; thirdly, when the main screen is ranked the highest, the display mode of the current display interface is maintained; when the sub-screen is ranked the highest, the sub-screen with the highest ranking is recommended to the user in the first way; in this way, different display modes can be determined based on the ranking of each screen, so that the most exciting video to be played can be presented to the user in a friendly manner, which solves the problem that the user misses the wonderful moments of the sub-screen program due to the relatively small screen of the sub-screen, thereby improving the user's experience of watching the program. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is one of the flowcharts of the video recommendation method according to an embodiment of the present application;

[0041] Figure 2 A schematic diagram of switching between the main screen and the secondary screen in an embodiment of the present application;

[0042] Figure 3 A schematic diagram of a frame of a highlighted flashing secondary screen according to an embodiment of the present application;

[0043] Figure 4 A schematic diagram of the structure of a video recommendation device according to an embodiment of the present application;

[0044] Figure 5 This is a second flowchart of the video recommendation method according to an embodiment of the present application;

[0045] Figure 6 A schematic diagram of the structure of a video recommendation device according to an embodiment of the present application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0047] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here. In addition, the "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally represents that the objects associated with each other are in an "or" relationship.

[0048] The video recommendation method, device, equipment, storage medium and program product provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings through specific embodiments and their application scenarios.

[0049] The embodiment of the present application provides a video recommendation method, such as Figure 1 As shown, the following steps are included:

[0050] Step 101, respectively acquiring characteristic data of each screen in the current display interface, wherein the screens obtained by splitting the current display interface include a main screen and at least one sub-screen; this step may acquire the above characteristic data based on a preset buried point;

[0051] Here, one, such as Figure 2 and Figure 3As shown, the size of the main screen is larger than that of the secondary screen; secondly, the feature data may include, for example, viewing popularity information and / or video information; specifically, the viewing popularity information may include, for example, one or more of: the number of viewing users, the number of barrage comments, and the number of likes; the video information is the information of the video itself played on each screen. Taking sports games as an example, the video information may include one or more of key moment information, star moment information, and game level information. Specifically, for example, the key moment information may include key stage information such as the deciding game, match point, overtime, etc. of the game, and the star moment information may include the status of star athletes, such as: name, popularity (such as world ranking, topical athlete, etc.), appearance time, etc.; the game level information may include championship games (games that produce champions, runners-up and third place) or promotion games; in addition, taking TV dramas as an example, the key moment information may include, for example, plot turning moments, etc., the star moment information may include, for example, the appearance time of characters played by actors with higher popularity, etc., and the video level information may be the rating level of the TV drama;

[0052] Step 102, sorting the main screen and the sub-screen according to the characteristic data. As an example, the sorting in this step can be based on the characteristic data of each screen to predict the excitement of the video content to be played on each screen, so as to sort the screens according to the predicted excitement. Taking sports games as an example, when it is determined according to the characteristic data (such as viewing popularity information) that the user's viewing enthusiasm increases and / or it is determined according to the characteristic data (such as video information) that the current time period is a match point or overtime or there is a star athlete appearing, the corresponding excitement should be increased;

[0053] Step 103a, when the ranking of the main screen is the highest, the display mode of the current display interface is maintained. Here, the ranking of the main screen is the highest, indicating that the video content to be played on the main screen is the most exciting; therefore, the display mode of the current display interface can be maintained, so that the most exciting content to be played can be played on the main screen with a larger size;

[0054] Step 103b, when the secondary screen has the highest ranking, the secondary screen with the highest ranking is recommended to the user in a first manner; similarly, the video content to be played on the secondary screen with the highest ranking is more exciting than the videos to be played on other secondary screens; in this way, the video content on the secondary screen with relatively high excitement can be recommended to the user based on the pre-set first manner, so as to achieve friendly video recommendation to the user.

[0055] Here, it should be noted that in order to further enhance the user's viewing experience, the video recommendation process is performed again after at least the first time period (such as 5 minutes) is executed after executing step 103b, that is, the video recommendation method of the embodiment of the present application can periodically recommend videos to the user to avoid frequently recommending the highest-ranked secondary screen through the first method and affecting the user's viewing effect.

[0056] In the video recommendation method of the embodiment of the present application, first, characteristic data of each screen in the current display interface is obtained respectively, wherein the screens obtained by splitting the current display interface include a main screen and at least one sub-screen; secondly, the main screen and the sub-screen are sorted according to the characteristic data; so as to determine the screen on which the most exciting video content is about to be played; thirdly, when the main screen is ranked highest, the display mode of the current display interface is maintained; when the sub-screen is ranked highest, the sub-screen with the highest ranking is recommended to the user in the first way; in this way, different display modes can be determined based on the ranking of each screen, so that the most exciting video to be played can be presented to the user in a friendly manner, thereby solving the problem of users missing the exciting moments of the sub-screen program due to the relatively small screen of the sub-screen, and improving the user's viewing experience.

[0057] As an optional implementation, step 102 includes:

[0058] According to each of the characteristic data, the viewing growth rate score, characteristic data score and wonderfulness score of the screen corresponding to the characteristic data are obtained; here, as mentioned above, the characteristic data may include viewing popularity information and / or video information; wherein the viewing growth rate score can be obtained based on the viewing popularity information obtained twice adjacently, the characteristic data score can be obtained based on the video information, and the wonderfulness score can be obtained based on the viewing growth rate score and the characteristic data score; the implementation process of this step will be specifically described later;

[0059] The main screen and the sub-screen are sorted according to the viewing growth rate score, the characteristic data score and the wonderfulness score.

[0060] As a specific implementation, obtaining, according to the feature data, a viewing growth rate score, a feature data score, and an exciting degree score of a screen corresponding to the feature data includes:

[0061] According to the viewing heat information in the acquired two adjacent characteristic data, the viewing heat growth rate of the screen corresponding to the characteristic data is calculated; here, taking the viewing heat information including the number of viewing users, the number of bullet screen comments and the number of likes as an example, the implementation process of this step is explained: first, the third difference between the number of viewing users of two adjacent times, the fourth difference between the number of bullet screen comments of two adjacent times and the fifth difference between the number of likes of two adjacent times are calculated; secondly, according to the third difference, the fourth difference and the fifth difference, the viewing heat growth rate is determined; wherein, when determining the viewing heat growth rate, the first method is: to determine by looking up a table (a correspondence table of the third difference, the fourth difference, the fifth difference and the viewing heat growth rate); the second method is: to determine based on a preset formula (for example, weighted summation of the third difference, the fourth difference and the fifth difference, etc.); the third method is: to calculate the weighted sum of the third difference, the fourth difference and the fifth difference When at least one of the values ​​is greater than zero, the viewing heat growth rate is determined according to at least part of the difference values ​​greater than zero. For example, when the third difference value is greater than zero, the viewing heat growth rate is determined to be the third difference value. When the third difference value is less than or equal to zero and at least one of the fourth difference value and the fifth difference value is greater than zero, the viewing heat growth rate is determined to be the weighted sum of the fourth difference value and the fifth difference value (when both the fourth difference value and the fifth difference value are greater than zero), or the viewing heat growth rate is determined to be the fourth difference value (when the fourth difference value is greater than zero and the fifth difference value is less than or equal to zero) or the fifth difference value (when the fourth difference value is less than or equal to zero and the fifth difference value is greater than zero), or the viewing heat growth rate is determined to be a preset growth rate, wherein the preset growth rate is a fixed value, or the preset growth rate is related to the fourth difference value and the fifth difference, and specifically, the preset growth rate related to the fourth difference value and the fifth difference may be obtained by looking up a table;

[0062] The viewing growth rate score is calculated according to the viewing heat growth rate and the pre-configured first weight; specifically, the viewing growth rate score is the product of the viewing heat growth rate and the first weight; for example, this step can be expressed as: G(x)=α*y(t); wherein G(x) represents the viewing growth rate score, α represents the first weight, and y(t) represents the viewing heat growth rate;

[0063] Determining key moment feature data according to key moment information in the feature data, and determining star moment feature data according to star moment information in the feature data;

[0064] Here, when the key moment information indicates that the video information is about to have a key moment, the key moment feature data is recorded as a first value (e.g., 1); when the key moment information indicates that the video information will not have a key moment in the near future, the key moment feature data is recorded as a second value (e.g., 0); and taking sports games as an example, the key moment information is, for example, key stage information such as a deciding game, a match point, and an overtime game, then based on the key moment information and the current game situation, it is predicted whether a key moment is about to occur, such as: when entering a deciding game, a match point, or an overtime game, it is determined to be a key moment, otherwise it is not a key moment;

[0065] Taking sports games as an example, the process of determining the star moment feature data according to the star moment information is as follows: predict whether it is a star moment by the appearance time and popularity ranking of the athletes. For example, if a well-known athlete appears, it is considered to be a star moment, otherwise it is not a star moment; wherein, if it is a star moment, the star moment feature data is recorded as a first value (for example, 1), and if it is not a star moment, the star moment feature data is recorded as a second value (for example, 0);

[0066] According to the video level in the feature data, a second weight corresponding to the key moment feature data and a third weight corresponding to the star moment feature data are determined; here, when the video level is higher than a preset level, the second weight is greater than the third weight, and when the video level is lower than the preset level, the second weight is less than the third weight; also taking a sports game as an example, when the video level is expressed by a game level, when the game level indicates a promotion game, the third weight is greater than the second weight, and when the game level is a championship game, the second weight is greater than the third weight;

[0067] The feature data score is calculated according to the key moment feature data, the star moment feature data, the second weight and the third weight; specifically, this step is: based on the second weight and the third weight, the key moment feature data and the star moment feature data are weighted and summed to obtain the feature data score; for example, this step can be expressed as: T(x)=β*g(t)+γ*m(t), wherein T(x) represents the feature data score, β represents the second weight, γ represents the third weight, g(t) represents the key moment feature data, and m(t) represents the star moment feature data;

[0068] The wonderfulness score is calculated according to the viewing growth rate score and the characteristic data score. Specifically, in this step, the wonderfulness score is the sum of the viewing growth rate score and the characteristic data score; for example, this step can be represented by the formula J(x)=G(x)+T(x)=α*y(t)+β*g(t)+γ*m(t), where J(x) represents the score value corresponding to the wonderfulness.

[0069] Here, it should be noted that the first weight, the second weight and the third weight can be adjusted based on actual conditions.

[0070] In the above specific implementation method, the excitement score of the content to be played on the screen is comprehensively calculated based on the viewing popularity growth rate, the video level, the key moment feature data, the star moment feature data and the pre-configured weights, so that the calculated excitement score value is more objective and accurate, so as to recommend the most exciting videos to users in the future and enhance the user's viewing experience.

[0071] Here, it should be noted that, since the embodiment of the present application performs real-time calculations on multiple program streams, in order to improve computing efficiency and save computing costs, for each screen, when determining that the viewing popularity of the screen has increased based on the viewing popularity growth rate, the viewing growth rate score, feature data score and excitement score (an increase in viewing popularity indicates an increase in excitement) can be further calculated. Among them, the step of determining whether the viewing popularity of the screen has increased based on the viewing popularity growth rate is, for example: determining whether the viewing popularity has increased based on the instantaneous growth rate of the viewing popularity (such as the change in viewing popularity in adjacent one minute), for example, an instantaneous growth rate greater than 0 indicates that the viewing popularity has increased, and an instantaneous growth rate less than or equal to 0 indicates that the viewing popularity has not increased;

[0072] Here, taking the viewing popularity information including the number of viewing users, the number of barrage comments and the number of likes as an example, the above steps are explained: calculate the third difference between the currently collected number of viewing users and the adjacent previous number of viewing users, the fourth difference between the currently collected number of barrage comments and the adjacent previous number of barrage comments, and the fifth difference between the currently collected number of likes and the adjacent previous number of likes, where, when the third difference is greater than 0, or the third difference is less than or equal to 0 and the fourth difference and / or the fifth difference is greater than 0, it indicates that the viewing popularity has increased.

[0073] As another specific implementation, the sorting of the main screen and the secondary screen according to the viewing growth rate score, the feature data score and the wonderfulness score includes:

[0074] According to the wonderfulness score, the main screen and the secondary screen are sorted in descending order to obtain a first sorting result;

[0075] In the first sorting result, when the main screen has the highest sorting, determining the sorting result of the main screen and the slave screen as the first sorting result;

[0076] In the first sorting result, if the slave screen sorted before the main screen exists, a first slave screen set is acquired, wherein the first slave screen set includes the slave screen sorted before the main screen; that is, the first slave screen set consists of one or more slave screens with a higher degree of excitement than the main screen;

[0077] According to the viewing growth rate score and the feature data score, the respective slave screens in the first slave screen set and the main screen are sorted to obtain the sorting results of the main screen and the slave screens.

[0078] Specifically, sorting the slave screens in the first slave screen set and the main screen according to the viewing growth rate score and the feature data score to obtain the sorting results of the main screen and the slave screens includes:

[0079] Compare the viewing growth rate score and the feature data score of the main screen with those of each of the secondary screens in the first secondary screen set, respectively, to obtain a second secondary screen set, wherein the viewing growth rate score and the feature data score of each of the secondary screens in the second secondary screen set are both greater than the viewing growth rate score and the feature data score of the main screen;

[0080] In the case that the second secondary screen set is an empty set, the secondary screens and the main screen in the second secondary screen set are sorted in descending order according to the wonderfulness score to obtain a second sorting result, wherein the second sorting result is a sorting result of the main screen and the secondary screen;

[0081] In the case that the second slave screen set is an empty set, the sorting result of the main screen and the slave screen is determined to be the first sorting result.

[0082] That is to say, when the second sub-screen set includes at least one sub-screen, according to the sorting results of each screen in the second sub-screen and the main screen, step 103a is executed (in the sorting result, the main screen is ranked highest) or step 103b is executed (in the sorting result, there is a sub-screen ranked higher than the main screen).

[0083] As an optional implementation, step 103b includes at least one of the following:

[0084] In the case where the ranking result of the main screen and the sub-screen is the first ranking result, the sub-screen with the highest ranking is recommended to the user by highlighting and flashing the border of the sub-screen with the highest ranking; here, when the ranking result of the main screen and the sub-screen is the first ranking result, it indicates that there are sub-screens with a higher level of excitement than the main screen, but the viewing growth rate scores and / or feature data scores of these sub-screens (sub-screens with a higher level of excitement than the main screen) are lower than the viewing growth rate score and / or feature data score of the main screen. Therefore, the sub-screen with the highest ranking can be recommended to the user by highlighting and flashing the border of the sub-screen with the highest level of excitement (the sub-screen with the highest ranking) among these sub-screens; here, highlighting and flashing the target sub-screen can specifically be single-speed flashing, such as Figure 3 As shown, the slave screen 2 is determined as the target slave screen. Figure 3 In the example, the frame of the secondary screen 2 is bolded to indicate double-speed flashing;

[0085] In the case where the ranking results of the main screen and the sub-screen are the second ranking results, the sub-screen with the highest ranking is recommended to the user by automatically switching the main screen and the sub-screen with the highest ranking; here, when the ranking results of the main screen and the sub-screen are the second ranking results, it indicates that there are sub-screens whose excitement scores, viewing growth rate scores and feature data scores are all higher than those of the main screen, and the scores of these sub-screens from different dimensions are higher than those of the main screen, indicating that the video contents to be played on these sub-screens should be strongly recommended to the user. Therefore, in this case, the sub-screen with the highest ranking in the second ranking result can be recommended to the user by automatically switching the main screen and the sub-screen with the highest ranking in the second ranking result; here, an example of switching the main screen and the sub-screen with the highest ranking in the second ranking result (sub-screen 1) is as follows Figure 2 As shown, the video 2 played on the slave screen 1 is switched to the main screen 1 for playback, and at the same time, the video 1 played on the main screen is switched to the slave screen 1 for playback.

[0086] Here, the implementation process of the above two optional implementation methods is illustrated by example: assuming that the viewing growth rate score, feature data score, and wonderfulness score of the wonderful moments of the main screen are recorded as Z[G(x)], Z[T(x)], and Z[J(x)] respectively; the viewing growth rate score, feature data score, and wonderfulness score of the wonderful moments of the secondary screen are recorded as F[G(x)], F[T(x)], and F[J(x)]; and assuming that there are n secondary screens on the display interface (n>=1); then they are recorded as: {Fn[G(x)], Fn[T(x)], and Fn[J(x)]} respectively; on this basis, the implementation process of the above two optional implementation methods is:

[0087] First, sort the viewing growth rate scores, feature data scores, and excitement level scores of the main screen and each secondary screen respectively. Secondly, compare the excitement level scores of the main screen and the secondary screens. If F[J(x)] > Z[J(x)], record the number M of secondary screens that meet the condition. Secondly, compare the viewing growth rate scores and feature data scores of the main screen and the secondary screens that meet F[J(x)] > Z[J(x)]. Among them, when M = 1, if F[G(x)] > Z[G(x)] and F[T(x)] > F[T(x)], then switch the main and secondary screens; otherwise, give a single-speed flashing prompt. When M > 1, if there are secondary screens where F[G(x)] > Z[G(x)] and F[T(x)] > F[T(x)], record the number N (0 < N <= M) of secondary screens that meet the condition. Among them, if N = 1, switch the secondary screen that meets the condition with the main screen; if N > 1, compare the excitement level scores of the N secondary screens that meet the condition, and record the highest score as F(max)[J(x)], and switch the secondary screen corresponding to the F(max)[J(x)] condition with the main screen.

[0088] The video recommendation method provided by the embodiment of the present application, in the case of multi-channel co-viewing, automatically identifies and predicts the most exciting moment of multi-channel videos (matches) at a certain moment, compares the excitement levels, and automatically presents the most exciting moment / most exciting content at the same moment to the user on the most suitable screen for the first time, so as to improve the user's viewing experience and interest.

[0089] Next, in combination with Figure 4 , the implementation process of an example of the video recommendation method of the present application will be described. Among them, this example includes the following steps:

[0090] Step 1: Data collection: Collect viewing popularity data (corresponding to the aforementioned viewing popularity information), critical moment data (corresponding to the aforementioned critical moment information), star moment data (corresponding to the aforementioned star moment information), and game level data (corresponding to the aforementioned video level); specifically: for viewing popularity information: collect the number of viewing users, the number of bullet comments, the number of likes, etc. of the main and secondary screen sports games in real time; for critical moment information: collect information on critical stages such as the decisive game, game point, overtime, etc. of the main and secondary screen sports games; for star moment information: collect all star player situations in the main and secondary screen sports games, such as names, popularity (such as world rankings, popular athletes), playing time, and collect star player appearance information in real time, and calculate whether there are star players on the field in the current time period; for game level information: collect whether the main and secondary screen sports games produce champions / runners-up / third-place finishers;

[0091] Step 2: Data processing and analysis: Extract key feature data related to wonderful moments, such as user number growth rate (corresponding to the aforementioned viewing popularity growth rate), key moment feature data, star player feature data (corresponding to the aforementioned star moment feature data), and game level data (corresponding to the aforementioned video level); Specifically, user number growth rate y(t): Calculate the growth rate of the number of viewing users through the number of viewing users, the number of barrage comments, the number of likes, etc., where the growth rate is the instantaneous growth rate within one minute; key moment feature data g(t): Predict whether it is a key moment through the key stage information such as the deciding game, match point, overtime, etc. of the game; if it enters the final game, match point or overtime, it is a key moment, otherwise it is not a key moment. If it is a key moment, it is recorded as 1, otherwise it is recorded as 0; star moment feature data m(t): Predict whether it is a star moment through the athlete's appearance time and popularity ranking. If a well-known athlete appears, it is a star moment, otherwise it is not a star moment. If it is a star moment, it is recorded as 1, otherwise it is recorded as 0; Competition level data: if there is a champion, runner-up or third place, it is a championship match; if there is no champion, runner-up or third place, it is a promotion match;

[0092] Step 3: Calculate the wonderful moment score, use the key feature data in step 2 to calculate the wonderful moment score (corresponding to the score corresponding to the above-mentioned wonderful degree); specifically, take the game level data and user growth rate, key moment feature data and star moment feature data in step 2 as input, and output the viewing growth rate score G(x) (corresponding to the above-mentioned first score value), feature data score T(x) (corresponding to the above-mentioned second score value), and wonderful degree score J(x) (corresponding to the above-mentioned score corresponding to the wonderful degree); wherein, each specific score calculation formula is as follows:

[0093] G(x) = α*y(t);

[0094] T(x)=β*g(t)+γ*m(t);

[0095] J(x)=G(x)+T(x)=α*y(t)+β*g(t)+γ*m(t);

[0096] Among them, α, β, and γ are weight coefficients, which can be adjusted according to actual needs. However, the weight values ​​of β and γ will change according to the competition level data. When the competition level data is the promotion match, γ>β; when the competition level data is the championship match, γ<β;

[0097] Here, it should be noted that: since the calculation is performed on multiple program streams in real time, in order to improve the calculation efficiency and save the calculation cost, the key moment feature data g(t) and the star moment feature data m(t) are represented by only 1 and 0. The above specific score calculation is performed when y(t)>0, otherwise it is not calculated;

[0098] Step 4: Recommend and present the best recommended moments to the user; specifically, sort and analyze the scores of the game content on each screen to obtain the best recommended presentation method and the recommended scenario application that the user finally sees; here, as assumed in the previous example, the viewing growth rate score, feature data score, and excitement score of the main screen’s exciting moments are recorded as Z[G(x)], Z[T(x)], and Z[J(x)]; the viewing growth rate score, feature data score, and excitement score of the secondary screen’s exciting moments are recorded as F[G(x)], F[T(x)], and F[J(x)]; if there are n secondary screens, (n>=1). They are recorded as: {Fn[G(x)], Fn[T(x)], and Fn[J(x)]}; based on the above assumptions, the viewing growth rate score, feature data score, and excitement score of the main and secondary screens are sorted according to the output of the prediction algorithm in step three. Figure 4 Step 4 includes the following specific steps to implement different recommendation strategies: Figure 4 The steps of step 4 are as follows:

[0099] Step 1): Compare the wonderfulness scores of the main screen and the secondary screen; wherein:

[0100] If F[J(x)]>Z[J(x)], then record the number of sub-screens M that meet the condition and go to step 2);

[0101] Step 2): Compare the viewing growth rate score and feature data score of the main screen and the secondary screen that meets step 1); wherein:

[0102] When M=1:

[0103] If there is a sub-screen with F[G(x)]>Z[G(x)] and F[T(x)]>F[T(x)], the main and sub-screens will switch; otherwise, the single-speed flashing prompt will be displayed;

[0104] When M>1:

[0105] Compare the M secondary screens with the main screen, and record the number of secondary screens N (0 <= N <= M) that meet the conditions of F[G(x)]>Z[G(x)] and F[T(x)]>F[T(x)];

[0106] For example, if N=1, the secondary screen that meets the conditions will be switched with the main screen;

[0107] If N≠1, go to step 3);

[0108] Step 3): Compare the wonderfulness scores of the secondary screens that meet the conditions of M>1, N≠1 in step 2), and the highest score is F(max)[J(x)];

[0109] When N=0, the secondary screen corresponding to F(max)[J(x)] flashes as a prompt;

[0110] When N>1, the secondary screen corresponding to F(max)[J(x)] switches with the main screen;

[0111] Switch or recommend page presentation: According to the rating ranking, switch between the main and secondary screens (such as Figure 2 ), border flashing prompt (such as Figure 3 ) and other methods to automatically recommend the most exciting moments of the current moment to the audience.

[0112] It should be noted here that after an autonomous switch or prompt occurs, there will be a 5-minute interval before the next automatic switch or prompt is initiated.

[0113] The embodiment of the present application provides a video recommendation method, specifically a method for switching between main and secondary screens for multiple simultaneous viewings, which automatically recommends and displays exciting moments to the audience on a suitable display screen by collecting audience data (corresponding to the aforementioned viewing popularity information) and game information (corresponding to the aforementioned video information) in real time. Specifically: collect feature data, including data such as the number of viewing users, the number of comments, the number of likes, as well as key moment information, star player conditions, and game level information. Predict exciting content based on feature data, sort and analyze the scores of the game content on each screen, and derive the best recommendation presentation method, for example: compare the excitement scores of the main screen and the secondary screen. If the score of the secondary screen is higher than that of the main screen, compare the viewing growth rate score and feature data score of the main screen and the secondary screen that meets the conditions. If there is no secondary screen that meets the conditions, display the exciting content on the main screen, and flash the border of the secondary screen with the highest excitement at double speed to prompt it. If there are multiple secondary screens that meet the conditions, compare their excitement scores again, and select the one with the highest score for switching display.

[0114] The embodiment of the present application also provides a video recommendation device, such as Figure 5 As shown, the device comprises:

[0115] An acquisition module 501 is used to respectively acquire characteristic data of each screen in the current display interface, wherein the screens obtained by splitting the current display interface include a main screen and at least one sub-screen;

[0116] A sorting module 502, used for sorting the main screen and the secondary screen according to the characteristic data;

[0117] Display module 503a, used for maintaining the display mode of the current display interface when the main screen is ranked highest;

[0118] The recommendation module 503b is used to recommend the slave screen with the highest ranking to the user in a first manner when the slave screen has the highest ranking.

[0119] Optionally, the sorting module 502 includes:

[0120] An acquisition submodule, for acquiring, according to each of the feature data, a viewing growth rate score, a feature data score and a wonderfulness score of the screen corresponding to the feature data;

[0121] The sorting submodule is used to sort the main screen and the sub-screen according to the viewing growth rate score, the characteristic data score and the wonderfulness score.

[0122] Optionally, the acquisition submodule includes:

[0123] A first calculation unit, configured to calculate a viewing heat growth rate of the screen corresponding to the feature data according to viewing heat information in two adjacent feature data acquired;

[0124] A second calculation unit, configured to calculate the viewing growth rate score according to the viewing popularity growth rate and a pre-configured first weight;

[0125] A first determining unit, configured to determine key moment feature data according to key moment information in the feature data, and determine star moment feature data according to star moment information in the feature data;

[0126] A second determining unit, configured to determine, according to the video level in the feature data, a second weight corresponding to the key moment feature data and a third weight corresponding to the star moment feature data;

[0127] A third calculation unit, configured to calculate the feature data score according to the key moment feature data, the star moment feature data, the second weight, and the third weight;

[0128] The fourth calculation unit is used to calculate the excitement score according to the viewing growth rate score and the feature data score.

[0129] Optionally, the sorting submodule includes:

[0130] A sorting unit, configured to sort the main screen and the sub-screen in descending order according to the wonderfulness score to obtain a first sorting result;

[0131] a third determining unit, configured to, when the primary screen has the highest ranking in the first ranking results, determine that the ranking result of the primary screen and the secondary screen is the first ranking result;

[0132] an acquiring unit, configured to acquire a first slave screen set when the slave screen ranked before the main screen exists in the first sorting result, wherein the first slave screen set includes the slave screen ranked before the main screen;

[0133] A sorting unit is used to sort the respective secondary screens in the first secondary screen set and the main screen according to the viewing growth rate score and the feature data score, and obtain the sorting results of the main screen and the secondary screens.

[0134] Optionally, the sorting unit includes:

[0135] a comparing subunit, configured to compare the viewing growth rate score and the feature data score of the main screen with the viewing growth rate score and the feature data score of each of the secondary screens in the first secondary screen set, respectively, to obtain a second secondary screen set, wherein the viewing growth rate score and the feature data score of each of the secondary screens in the second secondary screen set are both greater than the viewing growth rate score and the feature data score of the main screen;

[0136] a sorting subunit, configured to sort the secondary screens and the primary screen in the second secondary screen set in descending order according to the wonderfulness score, if the second secondary screen set is not an empty set, to obtain a second sorting result, wherein the second sorting result is a sorting result of the primary screen and the secondary screen;

[0137] The determining subunit is used to determine, when the second slave screen set is an empty set, that the sorting result of the main screen and the slave screen is the first sorting result.

[0138] Optionally, the recommendation module 503b is specifically configured to perform at least one of the following steps:

[0139] When the ranking result of the main screen and the slave screen is the first ranking result, the slave screen with the highest ranking is recommended to the user by highlighting and flashing the frame of the slave screen with the highest ranking;

[0140] When the ranking result of the main screen and the slave screen is the second ranking result, the slave screen with the highest ranking is recommended to the user by automatically switching the main screen and the slave screen with the highest ranking.

[0141] It should be noted here that the above-mentioned video recommendation device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned video recommendation method embodiment, and can achieve the same technical effect. The parts and beneficial effects that are the same as those in the method embodiment will not be specifically described here.

[0142] An embodiment of the present application also provides a video recommendation device, including a transceiver 610, a processor 600, a memory 620, and a program stored in the memory 620 and executable on the processor 600; wherein the processor 600 implements the video recommendation method as described above when executing the program.

[0143] The transceiver 610 is used to receive and send data under the control of the processor 600 .

[0144] Among them, Figure 6 In the embodiment of the present invention, the bus architecture may include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 600 and memory represented by memory 620. The bus architecture may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are not further described herein. The bus interface provides an interface. The transceiver 610 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices over a transmission medium.

[0145] The processor 600 is responsible for managing the bus architecture and general processing, and the memory 620 can store data used by the processor 600 when performing operations.

[0146] It should be noted here that the above-mentioned video recommendation device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned video recommendation method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those of the method embodiment will not be described in detail here.

[0147] The embodiment of the present application also provides a readable storage medium, on which a program is stored, and when the program is executed by a processor, each process of the above-mentioned video recommendation method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0148] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course, by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), and includes a number of instructions for executing the methods described in each embodiment of the present application.

[0149] Therefore, an embodiment of the present application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps in the video recommendation method as described above and can achieve the same technical effect. To avoid repetition, they are not described here.

[0150] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0151] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles described in the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A video recommendation method, characterized in that: include: Respectively acquiring characteristic data of each screen in the current display interface, wherein the screens obtained by splitting the current display interface include a main screen and at least one sub-screen; sorting the main screen and the sub-screen according to the characteristic data; When the main screen has the highest ranking, maintaining the display mode of the current display interface; In the case where the slave screen has the highest ranking, the slave screen with the highest ranking is recommended to the user in a first manner.

2. The method according to claim 1, characterized in that The sorting of the main screen and the slave screen according to the characteristic data includes: According to each of the feature data, obtaining a viewing growth rate score, a feature data score and a wonderfulness score of the screen corresponding to the feature data; The main screen and the sub-screen are sorted according to the viewing growth rate score, the characteristic data score and the wonderfulness score.

3. The method according to claim 2, characterized in that The step of obtaining, according to the feature data, a viewing growth rate score, a feature data score, and a wonderfulness score of a screen corresponding to the feature data includes: Calculating the viewing popularity growth rate of the screen corresponding to the feature data according to the viewing popularity information in the two adjacent feature data obtained; Calculate the viewing growth rate score according to the viewing popularity growth rate and a pre-configured first weight; Determining key moment feature data according to key moment information in the feature data, and determining star moment feature data according to star moment information in the feature data; Determining, according to the video level in the feature data, a second weight corresponding to the key moment feature data and a third weight corresponding to the star moment feature data; Calculating the feature data score according to the key moment feature data, the star moment feature data, the second weight, and the third weight; The excitement score is calculated based on the viewing growth rate score and the feature data score.

4. The method according to claim 2, characterized in that: The step of sorting the main screen and the auxiliary screen according to the viewing growth rate score, the feature data score and the wonderfulness score includes: According to the wonderfulness score, the main screen and the secondary screen are sorted in descending order to obtain a first sorting result; In the first sorting result, when the main screen has the highest sorting, determining the sorting result of the main screen and the slave screen as the first sorting result; In the first sorting result, if the slave screen sorted before the main screen exists, acquiring a first slave screen set, the first slave screen set including the slave screen sorted before the main screen; According to the viewing growth rate score and the feature data score, the respective slave screens in the first slave screen set and the main screen are sorted to obtain the sorting results of the main screen and the slave screens.

5. The method according to claim 4, characterized in that According to the viewing growth rate score and the feature data score, sorting the slave screens in the first slave screen set and the main screen to obtain the sorting results of the main screen and the slave screens includes: Compare the viewing growth rate score and the feature data score of the main screen with those of each of the secondary screens in the first secondary screen set, respectively, to obtain a second secondary screen set, wherein the viewing growth rate score and the feature data score of each of the secondary screens in the second secondary screen set are both greater than the viewing growth rate score and the feature data score of the main screen; In the case that the second secondary screen set is not an empty set, the secondary screens and the main screen in the second secondary screen set are sorted in descending order according to the wonderfulness score to obtain a second sorting result, wherein the second sorting result is a sorting result of the main screen and the secondary screen; In the case that the second slave screen set is an empty set, the sorting result of the main screen and the slave screen is determined to be the first sorting result.

6. The method according to claim 5, characterized in that Recommending the slave screen with the highest ranking to the user in a first manner includes at least one of the following: When the ranking result of the main screen and the slave screen is the first ranking result, the slave screen with the highest ranking is recommended to the user by highlighting and flashing the frame of the slave screen with the highest ranking; When the ranking result of the main screen and the slave screen is the second ranking result, the slave screen with the highest ranking is recommended to the user by automatically switching the main screen and the slave screen with the highest ranking.

7. A video recommendation device, characterized in that: include: An acquisition module, used to respectively acquire characteristic data of each screen in the current display interface, wherein the screens obtained by splitting the current display interface include a main screen and at least one sub-screen; A sorting module, used for sorting the main screen and the sub-screen according to the characteristic data; A display module, used for maintaining the display mode of the current display interface when the main screen is ranked highest; The recommendation module is used to recommend the slave screen with the highest ranking to the user in a first manner when the slave screen has the highest ranking.

8. A video recommendation device, characterized in that: include: A transceiver, a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the video recommendation method according to any one of claims 1 to 6 when executed by the processor.

9. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the video recommendation method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the video recommendation method according to any one of claims 1 to 6.

Images