Short video advertisement recommendation method and system

Abstract

The invention belongs to the field of internet advertisement services, and provides a short video advertisement recommendation method, which comprises the following steps of: capturing real-time interaction data of a user; obtaining a user tag according to the user real-time interaction data; obtaining an advertisement sequence according to the user label, and calculating an advertisement real-time concurrence efficiency index; and generating an optimal advertisement sequence combination according to the real-time concurrent efficiency index, and performing advertisement recommendation according to the optimal advertisement sequence combination. According to the advertisement recommendation method provided by the embodiment of the invention, the timeliness and the concurrent load of the advertisement can be comprehensively evaluated, so that the advertisement putting effect and efficiency are improved, the stability of a digital content multi-network channel and the user experience are enhanced, and the advertisement putting process is comprehensively optimized.

Description

Technical Field

[0001] The present invention belongs to the field of Internet advertising services, and particularly relates to a short video advertising recommendation method and system. Background Art

[0002] In the short video ecosystem, user interests show a high degree of dynamic changes. Affected by various factors such as hot events, social dissemination, and time-sensitive content, they exhibit frequent mutation characteristics. With the rapid development of short video platforms, the focus of users' interests quickly shifts with factors such as news hotspots and content dissemination on social media, making it difficult for traditional recommendation systems to accurately capture these sudden interest changes. Most traditional recommendation algorithms rely on static historical portraits, such as weekly / monthly behavior statistics, users' historical browsing records, etc. to construct user portraits. Although these static data reflect users' long-term interests to a certain extent, in the short video ecosystem, in the face of sudden hotspot events, traditional recommendation systems often cannot respond to the sudden changes in users' interests in a timely manner due to signal lag, resulting in the recommendation results being unable to accurately meet users' current interest needs.

[0003] As described in "An Advertising Placement Method, Intelligent Terminal and Computer Readable Storage Medium" with patent number CN114845170A, which conducts advertising recommendations based on static historical user portraits. This method relies on users' past behavior data. In the rapidly changing short video ecosystem, the focus of users' interests may have changed, and existing Internet advertising services are difficult to adapt to such highly dynamic interest changes, resulting in inefficient and inaccurate advertising placements. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems in the related art to some extent. To this end, the first object of the present invention is to propose a short video advertising recommendation method, which can comprehensively evaluate the timeliness and concurrent load of advertisements, not only improve the effect and efficiency of advertising placements, but also enhance the stability of multi-network channels of digital content and user experience, and comprehensively optimize the advertising placement process;

[0005] The second object of the present invention is to propose a short video advertising recommendation system.

[0006] To achieve the above object, the first aspect embodiment of the present invention proposes a short video advertising recommendation method, and the method includes the following steps:

[0007] S100, capture real-time user interaction data;

[0008] S200, obtain user tags according to the real-time user interaction data;

[0009] S300. Obtain the advertisement sequence according to the user tags, and calculate the real-time concurrent performance index of the advertisement;

[0010] S400. Generate the optimal advertisement sequence combination according to the real-time concurrent performance index, and perform advertisement recommendation according to the optimal advertisement sequence combination.

[0011] According to the advertisement recommendation method of the embodiment of the present invention, by comprehensively evaluating the timeliness and concurrent load of the advertisement, not only the effect and efficiency of advertisement placement can be improved, but also the stability of the multi-network channels of digital content and the user experience can be enhanced, and the advertisement placement process can be comprehensively optimized.

[0012] Further, in step S100, the user real-time interaction data includes: playback completion rate, video duration, like behavior or share behavior.

[0013] Record the playback completion rate, video duration, like behavior or share behavior times of the i-th video viewed by the user within the last 24 hours; use pec(i) to represent the playback completion rate of the i-th video viewed by the user within the last 24 hours, use time(i) to represent the video duration of the i-th video viewed by the user within the last 24 hours, where i is the serial number of the video, use K(i) to represent the like behavior or share behavior times of the i-th video viewed by the user within the last 24 hours, where i is the serial number of the video i = 1, 2,..., K, K represents the number of videos, and record the total duration of the videos viewed by the user within the last 24 hours as the total duration TZ, record the maximum value of the video durations of the videos viewed by the user within the last 24 hours as Tmax, record the minimum value of the video durations of the videos viewed by the user within the last 24 hours as Tmin, the average value of the video durations of the videos viewed by the user within the last 24 hours as TK, and record the average value of the like behavior or share behavior times of the videos viewed by the user within the last 24 hours as OCK.

[0014] In the short video ecosystem, the user interest shows high-frequency mutation characteristics affected by hot events, social dissemination, and time-sensitive content. Among them, the high-frequency mutation characteristic means that the user interest changes rapidly, violently and discontinuously due to external or internal factors in a short time; while the traditional recommendation system relies on static historical portraits such as weekly / monthly behavior statistics. For example, the advertisement placement method proposed in the patent No. CN114845170A, titled "An Advertisement Placement Method, Intelligent Terminal and Computer Readable Storage Medium", this advertisement placement method often has problems that the historical behavior cannot reflect the user's current interest focus due to signal lag when the attention is transferred caused by breaking news, and it is difficult to establish an effective portrait for new users or low-active users due to sparse data; to solve the above problems, the present invention proposes step S200;

[0015] Further, in step S200, the steps of obtaining user tags based on real-time user interaction data include:

[0016] S201, calculating the shortest effective preference duration TPHa and the longest effective preference duration TPHb according to the real-time user interaction data;

[0017] Denote the interaction adjustment factor KUI as K(i) / (1 + OCK + K(i)), and denote the effective video duration TOI as Calculate the first offset duration difference, where the first offset duration difference TD1 is Tmax * TOI / TZ; the second offset duration difference TD2 is Tmin * TOI / TZ; the shortest effective preference duration TPHa is TK - TD1, and the longest effective preference duration TPHb is TK + TD2, obtaining the preference window [TPHa, TPHb];

[0018] Among them, the effective video duration TOI is used to measure the user's patience level; [TPHa, TPHb] is the preference window. When the user's patience is low (TOI is small), the preference window shrinks towards short durations, and vice versa, for adaptively adjusting the curvature to obtain the accurate preference video range of the user and filtering out videos that are significantly inconsistent with the user's preferences.

[0019] S202, screening out videos with video durations between the shortest effective preference duration TPHa and the longest effective preference duration TPHb according to TPHa and TPHb, and denoting them as preference videos;

[0020] S203, obtaining user tags according to the video tags of the preference videos;

[0021] Obtain the video tag with the highest frequency of occurrence in the preference videos, and match the user tags to which the tag belongs to obtain the matching user tags.

[0022] The beneficial effect of this step is that: using the effective interest duration TOI as the Riemannian measure of the user's patience level, and constructing a dynamic preference window [TPHa, TPHb] in the manifold space through the coupling calculation of the video duration extreme value and the total duration TZ to obtain curvature adaptive adjustment, enabling the duration preference interval to automatically expand and contract with the content type and usage scenario, and still maintaining the portrait accuracy rate in the case of sparse data.

[0023] Further, in step S300, obtaining the advertisement sequence according to the user tags and calculating the advertisement real-time concurrency efficiency index includes:

[0024] S310, after screening out the matching advertisements from the database through the user's user tags, calculating the time-limited load time through the waiting time for the server to send the matching advertisements to other users in the last 1 hour;

[0025] The method for calculating the timeliness load time by the waiting time of the server sending the matched advertisement to other users in the most recent 1 hour is as follows:

[0026] S311, use ty(j) to represent the waiting time for the j-th matched advertisement sent to other users in the most recent 1 hour, where j represents the serial number of each matched advertisement, j = 1, 2, …, G, and G represents the number of matched advertisements; obtain the waiting time after receiving the user request for each matched advertisement in the most recent 1 hour; denote the maximum value of the waiting time for the j-th matched advertisement sent to other users in the most recent 1 hour as tmax, the average value of the waiting time for the matched advertisements sent to other users in the most recent 1 hour as tpj, and the median of the waiting time for the matched advertisements sent to other users in the most recent 1 hour as tav;

[0027] S312, calculate the timeliness load time fxt(j) of the j-th advertisement; where fxt(j) is ty(j) plus the difference ratio between the maximum waiting time and the average waiting time, and the difference ratio between the maximum waiting time and the average waiting time is the difference between tmax and tpj multiplied by The ratio with G × tav.

[0028] Among them, the timeliness load time fxt(j) can reflect the difference between the maximum waiting time and the average waiting time, and quantify the response performance of the advertisement in terms of time.

[0029] S320, obtain the concurrent user number data of the matched advertisement, where the concurrent user number data includes the maximum concurrent user number, the current concurrent user number, and the total concurrent user number within the most recent 1 hour, and denote the maximum concurrent user number of the j-th matched advertisement as bfsx(j), that is, the maximum number of users that the advertisement can support for simultaneous sending; the current concurrent user number of the j-th advertisement is denoted as bfs(j), that is, the number of users currently recommending this advertisement; the total concurrent user number of the j-th advertisement within the most recent 1 hour is bfsp(j), that is, the number of times this advertisement has been sent to users in the past hour;

[0030] S330, calculate the real-time concurrent efficiency index tfh(j) through the concurrent user number data and the timeliness load time: tfh(j) is fxt(j) multiplied by the ratio of the current total concurrent delay OCK1 and the total concurrent delay OCK2 in the most recent hour, where the current total concurrent delay OCK1 is OCK2 is G × tpj × at; where a is the current concurrent rate, a = bfs(j) / bfsx(j), and at is the total concurrent rate in the most recent hour, at = bfsp(j) / bfs(j);

[0031] Among them, the real-time concurrent efficiency index tfh(j) is used to measure the real-time response efficiency of an advertisement by combining the time-effect load time fxt(j) and the ratio of concurrent delay.

[0032] The beneficial effects of this step are as follows: By analyzing the concurrent load situation of advertisements, the recommendation of advertisements is efficiently managed, avoiding system crashes or response delays caused by excessive concurrent loads, and ensuring the stability and smoothness of advertisement delivery. In addition, by reducing the waiting delay of advertisements and improving timeliness through this step and visualizing and quantifying with the current real-time concurrent efficiency index, advertisers can adjust the delivery strategy according to the performance data of the current real-time concurrent efficiency index, optimize budget allocation, and further improve the return on investment of advertisements.

[0033] Furthermore, in step S400, generating an optimal advertisement sequence combination according to the real-time concurrent efficiency index and performing advertisement recommendation according to the optimal advertisement sequence combination includes:

[0034] Calculating the average value of all real-time concurrent efficiency indexes and denoting it as pjfh, denoting the advertisements with real-time concurrent efficiency indexes less than pjfh as preferred advertisements, and performing advertisement recommendation in ascending order of the real-time concurrent efficiency indexes of the preferred advertisements.

[0035] The beneficial effects of the present invention: By comprehensively evaluating the timeliness and concurrent load of advertisements, not only the effect and efficiency of advertisement delivery are improved, but also the stability of the multi-network channels of digital content and the user experience are enhanced, and the advertisement delivery process is comprehensively optimized.

[0036] To achieve the above object, an embodiment of the second aspect of the present invention further proposes a short-video advertisement recommendation system, where the short-video advertisement recommendation system includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in a short-video advertisement recommendation method, and the advertisement recommendation system of the short-video advertisement runs on computing devices such as satellites, desktop computers, laptops, palm computers, and cloud data centers.

[0037] By executing the short-video advertisement recommendation method through the short-video advertisement recommendation system, the effect and efficiency of advertisement delivery can be improved by comprehensively evaluating the timeliness and concurrent load of advertisements, and the stability of the multi-network channels of digital content and the user experience can also be enhanced, and the advertisement delivery process can be comprehensively optimized. Brief Description of the Drawings

[0038] Figure 1 Shown is a flowchart of a short-video advertisement recommendation method;

[0039] Figure 2 Shown is a structure diagram of a short-video advertisement recommendation system. Detailed Embodiment

[0040] Embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are intended to explain the present invention, and should not be construed as limiting the present invention.

[0041] Figure 1 Shown is a flowchart of a short video advertisement recommendation method.

[0042] Referring to Figure 1 , the present invention provides a short video advertisement recommendation method, and the method includes the following steps:

[0043] S100, capturing real-time interaction data of the user;

[0044] S200, obtaining user tags according to the real-time interaction data of the user;

[0045] S300, obtaining an advertisement sequence according to the user tags, and calculating the real-time concurrency efficiency index of the advertisement;

[0046] S400, generating an optimal advertisement sequence combination according to the real-time concurrency efficiency index, and performing advertisement recommendation according to the optimal advertisement sequence combination.

[0047] According to the advertisement recommendation method of the embodiments of the present invention, it is possible to.

[0048] Further, in step S100, the real-time interaction data of the user includes: play completion rate, video duration, like behavior or share behavior.

[0049] Record the play completion rate, video duration, like behavior or share behavior times of the i-th video viewed by the user within the most recent 24 hours; use pec(i) to represent the play completion rate of the i-th video viewed by the user within the most recent 24 hours, use time(i) to represent the video duration of the i-th video viewed by the user within the most recent 24 hours, where i is the serial number of the video, use K(i) to represent the like behavior or share behavior times of the i-th video viewed by the user within the most recent 24 hours, where i is the serial number of the video, i = 1, 2,..., K, K represents the number of videos, and denote the total duration of the videos viewed by the user within the most recent 24 hours as the total duration TZ, denote the maximum value of the video durations of the videos viewed by the user within the most recent 24 hours as Tmax, denote the minimum value of the video durations of the videos viewed by the user within the most recent 24 hours as Tmin, denote the average value of the video durations of the videos viewed by the user within the most recent 24 hours as TK, and denote the average value of the like behavior or share behavior times of the videos viewed by the user within the most recent 24 hours as OCK.

[0050] In the short video ecosystem, due to the influence of hot events, social dissemination, and time-sensitive content, users' interests exhibit high-frequency mutation characteristics. Traditional recommendation systems rely on static historical portraits, such as weekly / monthly behavior statistics. For example, in the patent with patent number CN114845170A and title "An Advertising Placement Method, Intelligent Terminal, and Computer-Readable Storage Medium", the proposed advertising placement method often has problems. When attention shifts caused by breaking news, historical behavior cannot reflect the user's current interest focus due to signal lag, and it is difficult to establish an effective portrait for new users or low-active users due to sparse data. To solve the above problems, step S200 is proposed in the present invention;

[0051] Further, in step S200, the steps of obtaining user tags based on real-time user interaction data include:

[0052] S201, calculate the shortest effective preference duration TPHa and the longest effective preference duration TPHb according to the real-time user interaction data;

[0053] Denote the interaction adjustment factor KUI as K(i) / (1 + OCK + K(i)), and denote the effective video duration TOI as Calculate the first offset duration difference. Among them, the first offset duration difference TD1 is Tmax * TOI / TZ; the second offset duration difference TD2 is Tmin * TOI / TZ; the shortest effective preference duration TPHa is TK - TD1, and the longest effective preference duration TPHb is TK + TD2, obtaining the preference window [TPHa, TPHb];

[0054] Among them, the effective video duration TOI is used to measure the user's patience level; [TPHa, TPHb] is the preference window. When the user's patience is low (TOI is small), the preference window shrinks towards short durations, and vice versa, for adaptive adjustment through curvature to obtain the accurate range of the user's preferred videos and filter out videos that are significantly inconsistent with the user's preferences.

[0055] S202, screen out the videos with video durations between the shortest effective preference duration TPHa and the longest effective preference duration TPHb according to the shortest effective preference duration TPHa and the longest effective preference duration TPHb, and denote them as preference videos;

[0056] S203, obtain user tags according to the video tags of the preference videos;

[0057] Obtain the video tag with the highest frequency of occurrence in the preference videos, and match the user tags to which the tag belongs to obtain the matching user tags.

[0058] S300, obtain the advertisement sequence according to the user tags and calculate the real-time concurrent efficiency index of the advertisement;

[0059] S310, after screening out matching advertisements from the database based on the user tags of the user, calculate the timeliness load time through the waiting time for the server to send the matching advertisements to other users in the most recent 1 hour;

[0060] The method for calculating the timeliness load time through the waiting time for the server to send the matching advertisements to other users in the most recent 1 hour is as follows:

[0061] S311, use ty(j) to represent the waiting time for the j-th matching advertisement in the most recent 1 hour to be sent to other users, where j represents the serial number of each matching advertisement, j = 1, 2,..., G, and G represents the number of matching advertisements; obtain the waiting time after receiving the user request for each matching advertisement in the most recent 1 hour; denote the maximum value of the waiting time for the j-th matching advertisement in the most recent 1 hour to be sent to other users as tmax, the average value of the waiting time for the matching advertisements in the most recent 1 hour to be sent to other users as tpj, and the median of the waiting time for the matching advertisements in the most recent 1 hour to be sent to other users as tav;

[0062] S312, calculate the timeliness load time fxt(j) of the j-th advertisement; among them, fxt(j) is ty(j) plus the differential ratio between the maximum waiting time and the average waiting time, where the differential ratio between the maximum waiting time and the average waiting time is the difference between tmax and tpj multiplied by the ratio with G × tav.

[0063] Among them, the timeliness load time fxt(j) can reflect the difference between the maximum waiting time and the average waiting time, and quantify the response performance of the advertisement in terms of time.

[0064] S320, obtain the concurrent user number data of the matching advertisements, where the concurrent user number data includes the maximum concurrent user number, the current concurrent user number, and the total concurrent user number within the most recent 1 hour, and denote the maximum concurrent user number of the j-th matching advertisement as bfsx(j), that is, the maximum number of users that the advertisement can support to be sent simultaneously; the current concurrent user number of the j-th advertisement is denoted as bfs(j), that is, the number of users currently recommending this advertisement; the total concurrent user number of the j-th advertisement within the most recent 1 hour is bfsp(j), that is, the number of users to whom this advertisement has been sent in the past hour;

[0065] S330, calculate the real-time concurrent efficiency index tfh(j) through the concurrent user number data and the timeliness load time: tfh(j) is fxt(j) multiplied by the ratio of the current total concurrent delay OCK1 and the total concurrent delay OCK2 in the most recent hour, where the current total concurrent delay OCK1 is OCK2 is G × tpj × at; where a is the current concurrency rate, a = bfs(j) / bfsx(j), and at is the total concurrency rate in the most recent hour, at = bfsp(j) / bfs(j);

[0066] Among them, the real-time concurrency efficiency index tfh(j) is used to measure the real-time response efficiency of advertisements by combining the ratio of the time limit load time fxt(j) and the concurrency delay.

[0067] S400, generate an optimal advertisement sequence combination according to the real-time concurrency efficiency index, and perform advertisement recommendation according to the optimal advertisement sequence combination;

[0068] Calculate the average value of all real-time concurrency efficiency indexes and denote it as pjfh. Denote the advertisements with real-time concurrency efficiency indexes less than pjfh as preferred advertisements, and perform advertisement recommendation in ascending order of the real-time concurrency efficiency indexes of the preferred advertisements. Figure 2 Shown is the structure diagram of the short video advertisement recommendation system.

[0069] Refer to Figure 2 Furthermore, the present invention also proposes a short video advertisement recommendation system 20. The short video advertisement recommendation system 20 includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in a short video advertisement recommendation method. The advertisement recommendation system 20 of the short video advertisement runs on computing devices such as satellites, desktop computers, laptops, handheld computers, and cloud data centers.

[0070] The advertisement recommendation system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it runs in the following units of the advertisement recommendation system:

[0071] The acquisition unit 21 is used to capture user real-time interaction data;

[0072] The conversion unit 22 is used to obtain user tags according to the user real-time interaction data;

[0073] The loading unit 23 is used to obtain an advertisement sequence according to the user tags and calculate the advertisement real-time concurrency efficiency index;

[0074] The management unit 24 is used to generate an optimal advertisement sequence combination according to the real-time concurrency efficiency index and perform advertisement recommendation according to the optimal advertisement sequence combination.

[0075] The described short video advertisement recommendation system can run on computing devices such as desktop computers, laptops, palmtop computers, and cloud servers. The described short video advertisement recommendation system, the advertisement recommendation system that can run may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above examples are merely examples of a short video advertisement recommendation system 20, and do not constitute a limitation on the short video advertisement recommendation system 20. It may include more or fewer components than the examples, or combine certain components, or different components. For example, the short video advertisement recommendation system may also include input / output devices, network access devices, buses, etc.

[0076] By executing the advertisement recommendation method for short video advertisements through the short video advertisement recommendation system 20, it is possible to comprehensively evaluate the timeliness and concurrent load of advertisements, not only improving the effect and efficiency of advertisement placement, but also enhancing the stability of multi-network channels for digital content and the user experience, and comprehensively optimizing the advertisement placement process.

[0077] It should be noted that the logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection parts with one or more wirings (electronic devices), portable computer disk cartridges (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber devices, and portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0078] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), and the like.

[0079] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0080] In the description of the present invention, it should be understood that the terms such as "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention.

[0081] In addition, the terms "first", "second", etc. used in the embodiments of the present invention are only for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly indicating the quantity of the technical features indicated in this embodiment. Thus, the features defined with the terms "first", "second", etc. in the embodiments of the present invention can clearly or implicitly indicate that at least one such feature is included in this embodiment. In the description of the present invention, the meaning of the word "plurality" is at least two or more than two, such as two, three, four, etc., unless otherwise specifically defined in the embodiments.

[0082] In the present invention, unless otherwise clearly specified or limited in the embodiments, the terms "installed", "connected", "coupled" and "fixed" etc. appearing in the embodiments shall be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or integrated. Understandably, it can also be a mechanical connection, an electrical connection, etc.; of course, it can also be directly connected, or indirectly connected through an intermediate medium, or it can be the communication inside two elements, or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific implementation situations.

[0083] In the present invention, unless otherwise clearly specified and limited, the first feature being "on" or "under" the second feature can be that the first and second features are in direct contact, or the first and second features are indirectly in contact through an intermediate medium. Moreover, the first feature being "above", "over" and "on top of" the second feature can be that the first feature is directly above or obliquely above the second feature, or merely indicates that the horizontal height of the first feature is higher than that of the second feature. The first feature being "under", "below" and "beneath" the second feature can be that the first feature is directly below or obliquely below the second feature, or merely indicates that the horizontal height of the first feature is lower than that of the second feature.

[0084] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A short video advertisement recommendation method, characterized in that, The method includes the following steps: S100, capturing real-time user interaction data; S200, obtaining user tags based on the real-time user interaction data; S300, obtaining an advertisement sequence based on the user tags and calculating the real-time concurrent performance index of the advertisement; S400, generating an optimal advertisement sequence combination based on the real-time concurrent performance index and performing advertisement recommendation according to the optimal advertisement sequence combination.

2. The short video advertisement recommendation method according to claim 1, wherein In step S100, the real-time user interaction data includes: play completion rate, video duration, like behavior or share behavior; pec(i) represents the play completion rate of the i-th video viewed by the user within the most recent 24 hours, time(i) represents the video duration of the i-th video viewed by the user within the most recent 24 hours, where i is the serial number of the video, and K(i) represents the number of like behaviors or share behaviors of the i-th video viewed by the user within the most recent 24 hours, where i is the serial number of the video, i = 1, 2,..., K, and K represents the number of videos.

3. A short video advertisement recommendation method according to claim 2, characterized in that In step S100, the total duration of the videos viewed by the user within the most recent 24 hours is denoted as the total duration TZ, the maximum value of the video durations of the videos viewed by the user within the most recent 24 hours is denoted as Tmax, the minimum value of the video durations of the videos viewed by the user within the most recent 24 hours is denoted as Tmin, the average value of the video durations of the videos viewed by the user within the most recent 24 hours is denoted as TK, and the average value of the number of like behaviors or share behaviors of the videos viewed by the user within the most recent 24 hours is denoted as OCK.

4. A short video advertisement recommendation method according to claim 1, characterized in that Step S200 includes: S201, calculating the shortest effective preference duration and the longest effective preference duration according to the real-time user interaction data; S202, screening out the videos with video durations between the shortest effective preference duration and the longest effective preference duration based on the shortest effective preference duration and the longest effective preference duration, and denoting them as preference videos; S203, obtaining user tags based on the video tags of the preference videos.

5. A short video advertisement recommendation method according to claim 1, characterized in that, Step S300 includes: S310, after screening out the matching advertisements from the database through the user tags of the user, calculating the time-limited load time by the waiting time for sending the matching advertisements to other users by the server within the most recent 1 hour; S320, obtaining the concurrent user number data of the matching advertisements; S330, calculating the real-time concurrent performance index through the concurrent user number data and the time-limited load time.

6. A short video advertisement recommendation system, characterized in that, The short video advertisement recommendation system includes: a processor, a memory, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of claims 1 to 5 in a short video advertisement recommendation method. The short video advertisement recommendation system runs on a desktop computer, a laptop computer, a handheld computer, or a cloud data center.

Images