Video recommendation method, device and computing equipment

Abstract

The embodiments of the present invention relate to the field of multimedia technology, and disclose a video recommendation method, apparatus, and computing device, the method comprising: obtaining a first video vector during a user's video viewing process; the first video vector is a type vector to which the video viewed by the user belongs; obtaining a recommended video based on the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video, or a tidbit short video of the video viewed by the user, and the second video vector is a type vector corresponding to the user's preferred video type; displaying identification information of the recommended video and a display label corresponding to the recommended video on a display interface, the display label being generated based on the duration of the recommended video. In the above manner, the embodiments of the present invention can provide personalized related recommendation services based on the user's cognition of the viewed video, the viewing environment, and the user's behavior, thereby improving the user experience.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of multimedia technology, and in particular to a video recommendation method, device and computing equipment. Background Art

[0002] When users watch videos on Migu Video, they will be recommended videos related to the videos they are watching. This type of scenario is collectively referred to as related recommendations. The key to related recommendations is to calculate the correlation between videos. Common methods include content-based content correlation calculation, item-based collaborative filtering algorithm, and model-based similarity calculation, such as Item2Vec. Content-based content correlation calculation is relatively intuitive. For example, two similar videos have the same video type, are directed by the same director, and are written by the same actor. At the same time, other forms of videos such as condensed and edited short videos are expanded for videos, which all belong to the category of similar videos. Similarity calculation based on collaborative filtering algorithm and model-based is obtained through training with machine learning algorithms and massive user behavior data. It expresses the statistical consistency of two contents and is a more profound correlation proved by user data behavior. This type of correlation does not necessarily mean that the videos are similar intuitively.

[0003] In existing related recommendation scenarios, the recommendation results are determined when the user starts watching the video. This recommendation method does not take into account the user's understanding of the video, their active behavior during the viewing process, and the current environment in which the video is being watched, and cannot provide users with personalized content display. Summary of the invention

[0004] In view of the above problems, embodiments of the present invention provide a video recommendation method, apparatus, and computing device, which overcome the above problems or at least partially solve the above problems.

[0005] According to one aspect of an embodiment of the present invention, a video recommendation method is provided, the method comprising: obtaining a first video vector during a process in which a user watches a video; the first video vector is a type vector to which the video watched by the user belongs; obtaining a recommended video based on the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, the second video vector is a type vector corresponding to the user's preferred video type, the related video is a video related to the video watched by the user calculated using a collaborative filtering algorithm, and the similar video is a video having the same elements as the video watched by the user; displaying identification information of the recommended video and a display tag corresponding to the recommended video on a display interface, the display tag being generated based on the length of the recommended video.

[0006] In an optional manner, obtaining a recommended video based on the first video vector and the second video vector includes: clustering the first video vector and the second video vector, and determining whether the first video vector and the second video vector are of the same type based on the clustering result; if the first video vector and the second video vector are of the same type, obtaining the related video from the related videos; if the first video vector and the second video vector are not of the same type, obtaining the viewing time of the user; and obtaining the recommended video based on the viewing time.

[0007] In an optional manner, obtaining the recommended video based on the viewing time includes: if the viewing time is less than a first time threshold, obtaining the recommended video from the similar videos and related videos of the user's historical viewing videos; if the viewing time exceeds the first time threshold and is less than the second time threshold, obtaining the recommended video from the similar videos, wherein the second time threshold is greater than the first time threshold; if the viewing time is greater than the second time threshold, obtaining the recommended video from the related videos.

[0008] In an optional manner, obtaining the recommended video according to the viewing time also includes: obtaining user behavior, the user behavior including at least one of switching clarity, dragging to fast forward, and speed; obtaining the recommended video from the related videos, the similar videos, the condensed short video or the behind-the-scenes short video according to the viewing time and the user behavior.

[0009] In an optional manner, the obtaining of the recommended video from the related videos, the similar videos, the condensed short video or the behind-the-scenes short video according to the viewing time and the user behavior includes: if the viewing time is less than or equal to a second time threshold and the user behavior indicates that the user likes the video watched by the user, then obtaining the recommended video from the related videos; if the viewing time is less than or equal to the second time threshold and the user behavior indicates that the user does not like the video watched by the user, then obtaining the recommended video from the condensed short video; if the viewing time is greater than the second time threshold and the user behavior of switching definition is detected during the playback of the video watched by the user, then obtaining the recommended video from the related videos and the original book related to the video watched by the user; if the viewing time is greater than the second time threshold and the user behavior of dragging to fast forward or speeding up is detected during the playback of the video watched by the user, or it is detected that the user watches until a preset time before the end of the video watched by the user, then obtaining the recommended video from the behind-the-scenes short video and the film review related to the video watched by the user.

[0010] In an optional manner, the method further includes: if the viewing time is less than or equal to the second time threshold, determining the time corresponding to when the video watched by the user is finished playing; if the time is within a preset time period, obtaining the recommended video from the condensed short video or the behind-the-scenes short video.

[0011] In an optional manner, the method further includes: generating a display label for characterizing the duration type according to the duration type of the recommended video; and when a user behavior of dragging a progress bar is detected, displaying the identification information of the recommended video, the display label corresponding to the recommended video, and a direction arrow on a display interface that plays the video watched by the user, wherein the direction arrow is used to indicate the direction of sliding from the dragging point on the progress bar to the identification information of the recommended video, and the display label is displayed on the identification information of the recommended video; if a user behavior of sliding from the progress bar to the identification information of any of the recommended videos along the sliding direction of the direction arrow is detected, the player is controlled to play the recommended video.

[0012] According to another aspect of an embodiment of the present invention, a video recommendation device is provided, the device comprising: a data acquisition unit, for acquiring a first video vector while a user is watching a video; the first video vector is a type vector to which the video watched by the user belongs; a video recommendation unit, for obtaining a recommended video based on the first video vector and a second video vector; the recommended video comprises at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, the second video vector is a type vector corresponding to a user's preferred video type, the related video is a video related to the video watched by the user calculated using a collaborative filtering algorithm, the similar video is a video having the same elements as the video watched by the user; a display unit, for displaying identification information of the recommended video and a display tag corresponding to the recommended video on a display interface, the display tag being generated based on the duration of the recommended video.

[0013] According to another aspect of an embodiment of the present invention, there is provided a computing device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0014] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the steps of the above-mentioned video recommendation method.

[0015] According to another aspect of the embodiments of the present invention, a computer storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables the processor to execute the steps of the above-mentioned video recommendation method.

[0016] The embodiment of the present invention obtains a first video vector while a user is watching a video; the first video vector is a type vector to which the video watched by the user belongs; a recommended video is obtained based on the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, and the second video vector is a type vector corresponding to the user's preferred video type; identification information of the recommended video and a display label corresponding to the recommended video are displayed on a display interface, the display label is generated based on the length of the recommended video, and can provide users with personalized related recommendation services based on the user's cognition of the video, the user's viewing environment and the user's behavior, thereby improving the user experience.

[0017] The above description is only an overview of the technical solution of the embodiment of the present invention. In order to more clearly understand the technical means of the embodiment of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0019] Figure 1 A schematic diagram showing a flow chart of a video recommendation method provided by an embodiment of the present invention;

[0020] Figure 2 A recommendation schematic diagram of a video recommendation method provided by an embodiment of the present invention is shown;

[0021] Figure 3 A schematic diagram showing a recommendation method for recommending a video in which the viewing time is less than a second time threshold according to an embodiment of the present invention;

[0022] Figure 4 A schematic diagram showing a recommendation method for recommending a video when the viewing time exceeds a second time threshold according to an embodiment of the present invention;

[0023] Figure 5 A schematic diagram showing display of recommended videos in a vertical screen mode of a video recommendation method provided by an embodiment of the present invention is shown;

[0024] Figure 6 A schematic diagram showing the display of recommended videos in full-screen mode of the video recommendation method provided by an embodiment of the present invention is shown;

[0025] Figure 7 A schematic diagram showing the structure of a video recommendation device provided by an embodiment of the present invention is shown;

[0026] Figure 8 A schematic diagram of the structure of a computing device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0027] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0028] Figure 1 FIG. 1 is a flow chart of a video recommendation method provided by an embodiment of the present invention. The video recommendation method is applied to a server side, such as Figure 1 As shown, the video recommendation method includes:

[0029] Step S11: while the user is watching a video, a first video vector is obtained; the first video vector is a type vector to which the video watched by the user belongs.

[0030] In an embodiment of the present invention, the first video vector is a type vector to which the video watched by the user belongs. The method for obtaining the first video vector is as follows: obtain the type of the video M1 watched by the user. If there are 11 types of videos in total, they are marked as vectors as follows: [Thriller: 0.2, Crime: 0.4, Suspense: 0.5], where the number is the type value obtained by quantifying the type of the video M1. The larger the number, the closer the video M1 is to the corresponding type. The corresponding vector is: [0, 0, 0.2, 0, 0, 0.4, 0, 0, 0.5, 0, 0] (the 2nd, 5th, and 9th bits of the vector). There are 11 types of videos, that is, the vector has 11 dimensions, and the vector is defined as the first video vector. The specific dimension of the first video vector can be set as needed and is not limited here. It can be understood that a video may have the characteristics of different types of videos at the same time. For example, a video may meet the characteristics of "thriller", "crime" and "suspense" type videos at the same time.

[0031] Step S12: Obtain recommended videos based on the first video vector and the second video vector; the recommended videos include at least one of related videos, similar videos, condensed short videos or behind-the-scenes short videos of the video watched by the user, the second video vector is a type vector corresponding to the user's preferred video type, the related videos are videos related to the video watched by the user calculated using a collaborative filtering algorithm, and the similar videos are videos that have the same elements as the video watched by the user.

[0032] In the embodiment of the present invention, during the process of the user watching the video M1, a second video vector is also obtained. The second video vector is a type vector corresponding to the user's preferred video type, which can be obtained by obtaining the user's preferred video type and marking it as a user portrait of the user's video type preference, and then determining the user's preference for the video type. Suppose the user's preferences are as follows: [comedy: 0.8, thriller: 0.2, war: 1.0, crime: 0.4], where the number is the preference type value obtained by quantifying the user's preference. The larger the number, the stronger the user's preference for the type; mapped to the video type vector, the expression in the same vector space as the first video vector of the type to which the video belongs can be obtained: [0.8, 0, 0.2, 0, 1.0, 0.4, 0, 0, 0, 0, 0] (the 0th, 2nd, 4th, and 5th positions of the vector), which is called the second video vector. Of course, the second video vector can be completed using any existing technology, and the specific portrait algorithm used in the embodiment of the present invention is not limited.

[0033] Before step S12, related videos and similar videos corresponding to the video watched by the user are also obtained. Related videos are videos related to the video watched by the user calculated using a collaborative filtering algorithm. They are obtained based on the correlation between videos calculated using a machine learning algorithm and user behavior data. This type of correlation expresses an intrinsic connection between videos that is displayed through user behavior, and is suitable for recommendation by users who have a certain understanding of the video. The method for obtaining related videos is: obtaining the user's video viewing sequence; using an item-based collaborative filtering algorithm to calculate a set of related videos for each video in the video viewing sequence.

[0034] Similar videos are videos that have the same elements as the video watched by the user. They are obtained by intuitively searching for videos that have the same elements as the video watched by the user through the elements of the video (such as video type, director, actors, etc.). The recommendation of similar videos is relatively intuitive and suitable for people who do not have a strong understanding of videos. The method for obtaining similar videos is as follows: obtain the second video vector of the video in the video viewing sequence; based on the second video vector, use the cosine angle calculation method to calculate the similarity between the video in the video viewing sequence and the video that can be recommended by the video server; based on the similarity, determine the similar video set, condensed short video and behind-the-scenes short video for each video in the video viewing sequence. Condensed short videos and behind-the-scenes short videos are extensions of the videos watched by the user, and are usually used for extended recommendations in specific scenarios.

[0035] In step S12, optionally, Figure 2 As shown, the user clicks to start watching video M1, and determines whether the current time is within the preset time period. If so, a recommended video is obtained from the condensed short video or the behind-the-scenes short video. Otherwise, the first video vector and the second video vector are subjected to the same type of judgment, and the recommended video is obtained according to the same type of judgment result. The preset time period can be set by the user as needed, preferably late at night, such as 24:00 and later.

[0036] When determining whether the first video vector and the second video vector are of the same type, optionally, clustering is performed on the first video vector and the second video vector, and determining whether the first video vector and the second video vector are of the same type based on the clustering result. Clustering refers to clustering the second video vectors of video type preferences of a large number of users including the current user, and the first video vectors of a large number of videos including the video M1 watched by the user. If the first video vector and the second video vector are clustered to obtain the same category, then the first video vector and the second video vector are determined to be of the same type; otherwise, the first video vector and the second video vector are of different types. If the first video vector and the second video vector are of the same type, the recommended video is obtained from the related videos. If the first video vector and the second video vector are not of the same type, the viewing time of the user is obtained; and the recommended video is obtained based on the viewing time.

[0037] If the first video vector and the second video vector are not of the same type, further, combining Figure 2-Figure 4If the viewing time is less than or equal to the first time threshold, the recommended video is obtained from the similar videos and the related videos of the user's historical viewing videos, so that the user has more choices of interest when he does not like the video. If the viewing time exceeds the first time threshold and is less than the second time threshold, the recommended video is obtained from the similar videos, where the second time threshold is greater than the first time threshold. If the viewing time is greater than the second time threshold, the recommended video is obtained from the related videos.

[0038] In the process of watching videos by users, it is also necessary to judge the active behavior of users, and pay attention to the switching of definition, dragging and fast forwarding, and speed during the process of watching videos. If the viewing time is less than the first time threshold, it proves that the user may not have a strong enough understanding of the video, and the recommended content is mainly similar videos, and the recommended video is obtained in combination with the user behavior. If the viewing time is greater than the first time threshold, it proves that the user has a sufficient understanding of the video. At this time, the recommendation of related videos is mainly used, and the video recommendation is also combined with the user behavior. Optionally, the user behavior is obtained, and the user behavior includes at least one of switching definition, dragging and fast forwarding, and speed. Specifically, the user's viewing scene and specific user behavior during viewing can be monitored, such as switching to high definition in a non-WIFI environment, switching to low definition in any network environment, and dragging and fast forwarding, speeding up, etc. during the video viewing process. The user behavior can include one or more of switching definition, dragging and fast forwarding, and speeding up at the same time. The recommended video is obtained from the related video, the similar video, the condensed short video, or the behind-the-scenes short video according to the viewing time and the user behavior.

[0039] In an embodiment of the present invention, optionally, in combination with Figure 2 and Figure 3 , if the viewing time is less than or equal to the second time threshold, and the user behavior indicates that the user likes the video the user is watching, the recommended video is obtained from the related videos. If the viewing time is less than or equal to the second time threshold, and the user behavior indicates that the user does not like the video the user is watching, the recommended video is obtained from the condensed short video, so that the user can quickly understand the movie. In an embodiment of the present invention, when the viewing time is less than or equal to the second time threshold, if a user behavior of switching to high definition in a non-WIFI environment or switching to low definition in any network environment is detected during the playback of the video the user is watching, it is determined that the user likes the video the user is watching. If the video is switched to low definition throughout the process (any network environment is acceptable), it proves that the user has the intention to continue watching the video regardless of the network environment, indicating that the user likes the video. If a user behavior of dragging to fast forward or speeding up is detected during the playback of the video the user is watching, it is determined that the user does not like the video the user is watching.

[0040] Combination Figure 2 and Figure 4 , if the viewing time is longer than the second time threshold, and the user behavior of switching definition is detected during the playback of the video watched by the user, the recommended video is obtained from the related videos and the original books related to the video watched by the user. When the viewing time is longer than the second time threshold, if the user behavior of switching to high definition in a non-WIFI environment or switching to low definition in any network environment is detected during the playback of the video watched by the user, it means that the user likes the current video very much and has sufficient knowledge of the current video, and the recommended video can be obtained from the related videos and the original books related to the video watched by the user for recommendation.

[0041] If the viewing time is greater than the second time threshold, and the user behavior of dragging fast forward or speeding up is detected during the playback of the video watched by the user, or it is detected that the user has watched until a preset time before the end of the video watched by the user, then the recommended video is obtained from the short video of the highlights and the film reviews related to the video watched by the user. If the user behavior of dragging fast forward or speeding up is detected during the playback of the video watched by the user, it means that the user does not dislike the video watched by the user, but just does not have enough time. The short videos of highlights and high-quality film reviews can allow the user to continue to understand the video watched by the user in the subsequent fragmented time. If it is detected that the user has watched until a preset time before the end of the video watched by the user, it means that the user will watch the video watched by the user in its entirety. Obtaining recommended videos from short video highlights and high-quality film reviews can allow the user to have a deeper understanding of the video watched by the user. The preset time can also be set by the user as needed, and the preset time is preferably 5 minutes.

[0042] In the embodiment of the present invention, the first duration threshold is smaller than the second duration threshold. The first duration threshold and the second duration threshold can be set by the user as needed. Preferably, the first duration threshold is 5% and the second duration threshold is 50%.

[0043] In the embodiment of the present invention, the first video vector and the second video vector are not of the same type. If the viewing time is less than or equal to the second time threshold, the time corresponding to when the video watched by the user is finished is also determined; if the time is within the preset time period, the recommended video is obtained from the condensed short video or the behind-the-scenes short video so that the user can finish watching it as soon as possible without staying up late. The preset time period is preferably 24:00 and later.

[0044] The embodiment of the present invention uses the first video vector and the second video vector to perform vector clustering to determine whether the first video vector and the second video vector are of the same type. If they are of the same type, the user's cognition of the video watched by the user is sufficient, and related videos can be recommended. If they are of different types, the user is new to such videos and needs to be recommended with caution. In the cautious recommendation process, varying recommendations are made to the user based on the user's viewing time. Related videos require stronger cognition of the video itself than similar videos before they can be recommended, so they are recommended when the viewing time is longer. In addition to viewing time, at a specific viewing time, the embodiment of the present invention also considers recommending videos in combination with user behavior, and recommending condensed short videos and behind-the-scenes short videos in combination with the current scene, so that the recommended videos are more in line with the user's potential viewing needs.

[0045] Step S13: Displaying identification information of the recommended video and a display label corresponding to the recommended video on a display interface, wherein the display label is generated according to the duration of the recommended video.

[0046] In the embodiment of the present invention, since short videos are introduced as recommendation objects, in order to facilitate users to understand whether the video recommended for users is a long video, a condensed short video or a behind-the-scenes short video, so as to facilitate video selection, a display method that facilitates users to select recommendation results can be applied. Optionally, a display label representing the duration type is generated according to the duration type of the recommended video; and when the user behavior of dragging the progress bar is detected, the identification information of the recommended video, the display label corresponding to the recommended video and the direction arrow are displayed on the display interface of the video being played by the user, the direction arrow is used to indicate the direction of sliding from the dragging position on the progress bar to the identification information of the recommended video, and the display label is displayed on the identification information of the recommended video. Among them, the duration type includes one of a full video, a condensed short video and a behind-the-scenes short video, and the identification information of the recommended video is preferably a poster of the recommended video. In order to facilitate users to understand whether the video recommended for the user is a long video, a condensed short video, or a short video with highlights, the embodiment of the present invention will display the identification information of the recommended video and the display label corresponding to the recommended video on the display interface of the video playback watched by the user according to the duration type of the recommended video after determining the recommended video, so as to inform the user of the duration type of the recommended video, thereby facilitating the user's selection. Figure 5 and Figure 6 As shown, Figure 5 Display recommended videos in portrait mode. Figure 6It is the recommended video display in full-screen mode. The circular frame in the figure shows the poster of the recommended video and the display label corresponding to the recommended video. For example: if the video watched by the user is "Seven Deadly Sins", the poster of the related video is located on the far left of the figure, such as "The Curious Case of Benjamin Button", the poster of the condensed short video and the display label corresponding to the condensed short video are displayed in the middle part of the figure, and the poster of the behind-the-scenes short video and the display label corresponding to the behind-the-scenes short video are displayed on the right side of the figure. In full-screen mode, the poster of the recommended video can be displayed with a certain degree of transparency to avoid excessive obstruction of the display interface of the video playback watched by the user. Of course, in full-screen mode, it is also possible not to make recommendations on the display interface. Whether to make recommendations can be turned off or on by the user through function selection.

[0047] If a user behavior of sliding from the progress bar to the identification information of any of the recommended videos along the sliding direction of the direction arrow is detected, the player is controlled to play the recommended video. For example, when the user drags the progress bar, a pop-up window such as Figure 5 or Figure 6 The display interface displays the identification information of the recommended video and the display label and direction arrow corresponding to the recommended video; if it is detected that the user's finger slides along the direction arrow to the poster of any recommended video in the figure, the player is controlled to play the recommended video. In this way, the two operations of dragging the progress bar and selecting the recommended video can be completed without the user's finger leaving the interface, which improves the efficiency of the user in selecting the recommended video when dragging the progress bar.

[0048] The embodiment of the present invention obtains a first video vector while a user is watching a video; the first video vector is a type vector to which the video watched by the user belongs; a recommended video is obtained based on the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, and the second video vector is a type vector corresponding to the user's preferred video type; identification information of the recommended video and a display label corresponding to the recommended video are displayed on a display interface, the display label is generated based on the length of the recommended video, and can provide users with personalized related recommendation services based on the user's cognition of the video, the user's viewing environment and the user's behavior, thereby improving the user experience.

[0049] Figure 7 FIG. 2 shows a schematic diagram of the structure of a video recommendation device according to an embodiment of the present invention. Figure 7 As shown, the video recommendation device includes: a data acquisition unit 701, a video recommendation unit 702 and a display unit 703. Among them:

[0050] The data acquisition unit 701 is used to obtain a first video vector while a user is watching a video; the first video vector is a type vector to which the video watched by the user belongs; the video recommendation unit 702 is used to obtain a recommended video based on the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, the second video vector is a type vector corresponding to the user's preferred video type, the related video is a video related to the video watched by the user calculated using a collaborative filtering algorithm, and the similar video is a video having the same elements as the video watched by the user; the display unit 703 is used to display the identification information of the recommended video and a display tag corresponding to the recommended video on a display interface, and the display tag is generated based on the length of the recommended video.

[0051] In an optional manner, the video recommendation unit 702 is used to: cluster the first video vector and the second video vector, and determine whether the first video vector and the second video vector are of the same type based on the clustering result; if the first video vector and the second video vector are of the same type, obtain the recommended video from the related videos; if the first video vector and the second video vector are not of the same type, obtain the user's viewing time; and obtain the recommended video based on the viewing time.

[0052] In an optional manner, the video recommendation unit 702 is used to: if the viewing time is less than a first time threshold, obtain the recommended video from the similar videos and related videos of the user's historical viewing videos; if the viewing time exceeds the first time threshold and is less than the second time threshold, obtain the recommended video from the similar videos, wherein the second time threshold is greater than the first time threshold; if the viewing time is greater than the second time threshold, obtain the recommended video from the related videos.

[0053] In an optional manner, the video recommendation unit 702 is used to: obtain user behavior, the user behavior including at least one of switching clarity, dragging to fast forward, and speed; obtain the recommended video from the related videos, the similar videos, the condensed short video or the behind-the-scenes short video according to the viewing time and the user behavior.

[0054] In an optional manner, the video recommendation unit 702 is used to: if the viewing time is less than or equal to a second time threshold, and the user behavior indicates that the user likes the video watched by the user, then obtain the recommended video from the related videos; if the viewing time is less than or equal to the second time threshold, and the user behavior indicates that the user does not like the video watched by the user, then obtain the recommended video from the condensed short video; if the viewing time is greater than the second time threshold, and the user behavior of switching clarity is detected during the playback of the video watched by the user, then obtain the recommended video from the related videos and the original book related to the video watched by the user; if the viewing time is greater than the second time threshold, and the user behavior of dragging to fast forward or speeding up is detected during the playback of the video watched by the user, or it is detected that the user watches to a preset time before the end of the video watched by the user, then obtain the recommended video from the behind-the-scenes short video and the film reviews related to the video watched by the user.

[0055] In an optional manner, the video recommendation unit 702 is used to: if the viewing time is less than or equal to the second time threshold, determine the time corresponding to when the video watched by the user is finished playing; if the time is within a preset time period, obtain the recommended video from the condensed short video or the behind-the-scenes short video.

[0056] In an optional manner, the display unit 703 is used to: generate a display label for characterizing the duration type according to the duration type of the recommended video; and when a user behavior of dragging a progress bar is detected, display the identification information of the recommended video, the display label corresponding to the recommended video, and a direction arrow on the display interface that plays the video watched by the user, wherein the direction arrow is used to indicate the direction of sliding from the dragging point on the progress bar to the identification information of the recommended video, and the display label is displayed on the identification information of the recommended video; if a user behavior of sliding from the progress bar to the identification information of any of the recommended videos along the sliding direction of the direction arrow is detected, the player is controlled to play the recommended video.

[0057] The embodiment of the present invention obtains a first video vector while a user is watching a video; the first video vector is a type vector to which the video watched by the user belongs; a recommended video is obtained based on the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, and the second video vector is a type vector corresponding to the user's preferred video type; identification information of the recommended video and a display label corresponding to the recommended video are displayed on a display interface, the display label is generated based on the length of the recommended video, and can provide users with personalized related recommendation services based on the user's cognition of the video, the user's viewing environment and the user's behavior, thereby improving the user experience.

[0058] An embodiment of the present invention provides a non-volatile computer storage medium, wherein the computer storage medium stores at least one executable instruction, and the computer executable instruction can execute the video recommendation method in any of the above method embodiments.

[0059] The executable instructions may be specifically used to cause the processor to perform the following operations:

[0060] In the process of a user watching a video, a first video vector is obtained; the first video vector is a type vector to which the video watched by the user belongs;

[0061] According to the first video vector and the second video vector, a recommended video is obtained; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, the second video vector is a type vector corresponding to the type of video preferred by the user, the related video is a video related to the video watched by the user calculated using a collaborative filtering algorithm, and the similar video is a video having the same elements as the video watched by the user;

[0062] The identification information of the recommended video and a display tag corresponding to the recommended video are displayed on a display interface, wherein the display tag is generated according to the duration of the recommended video.

[0063] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0064] Clustering the first video vector and the second video vector, and determining whether the first video vector and the second video vector are of the same type according to the clustering result;

[0065] If the first video vector and the second video vector are of the same type, obtaining the recommended video from the related videos;

[0066] If the first video vector and the second video vector are not of the same type, obtaining the user's viewing time; and obtaining the recommended video according to the viewing time.

[0067] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0068] If the viewing time is less than a first time threshold, obtaining the recommended video from the similar videos and related videos of the user's historical viewing videos;

[0069] If the viewing time exceeds the first time threshold and is less than the second time threshold, obtaining the recommended video from the similar videos, wherein the second time threshold is greater than the first time threshold;

[0070] If the viewing time is greater than the second time threshold, the recommended video is obtained from the related videos.

[0071] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0072] Acquiring a user behavior, wherein the user behavior includes at least one of switching definition, dragging to fast forward, and multiple speed;

[0073] The recommended video is obtained from the related videos, the similar videos, the condensed short videos or the behind-the-scenes short videos according to the viewing time and the user behavior.

[0074] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0075] If the viewing time is less than or equal to a second time threshold, and the user behavior indicates that the user likes the video the user is watching, obtaining the recommended video from the related videos;

[0076] If the viewing time is less than or equal to the second time threshold, and the user behavior indicates that the user does not like the video the user is watching, obtaining the recommended video from the condensed short video;

[0077] If the viewing time is greater than the second time threshold, and the user behavior of switching definition is detected during the playback of the video watched by the user, obtaining the recommended video from the related videos and the original books related to the video watched by the user;

[0078] If the viewing time is greater than the second time threshold, and the user behavior of dragging to fast forward or double the speed is detected during the playback of the video watched by the user, or it is detected that the user watches until a preset time before the end of the video watched by the user, then the recommended video is obtained from the behind-the-scenes short video and the film reviews related to the video watched by the user.

[0079] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0080] If the viewing time is less than or equal to the second time threshold, determining a time corresponding to when the video viewed by the user is finished playing;

[0081] If the moment is within a preset time period, the recommended video is obtained from the condensed short video or the behind-the-scenes short video.

[0082] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0083] Generate a display label for characterizing the duration type according to the duration type of the recommended video; and when detecting the user behavior of dragging the progress bar, display the identification information of the recommended video, the display label corresponding to the recommended video, and a direction arrow on the display interface playing the video watched by the user, wherein the direction arrow is used to indicate the direction of sliding from the dragging position on the progress bar to the identification information of the recommended video, and the display label is displayed on the identification information of the recommended video;

[0084] If a user behavior of sliding from the progress bar to the identification information of any of the recommended videos along the sliding direction of the direction arrow is detected, the player is controlled to play the recommended video.

[0085] The embodiment of the present invention obtains a first video vector while a user is watching a video; the first video vector is a type vector to which the video watched by the user belongs; a recommended video is obtained based on the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, and the second video vector is a type vector corresponding to the user's preferred video type; identification information of the recommended video and a display label corresponding to the recommended video are displayed on a display interface, the display label is generated based on the length of the recommended video, and can provide users with personalized related recommendation services based on the user's cognition of the video, the user's viewing environment and the user's behavior, thereby improving the user experience.

[0086] An embodiment of the present invention provides a computer program product, which includes a computer program stored on a computer storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the video recommendation method in any of the above method embodiments.

[0087] The executable instructions may be specifically used to cause the processor to perform the following operations:

[0088] In the process of a user watching a video, a first video vector is obtained; the first video vector is a type vector to which the video watched by the user belongs;

[0089] According to the first video vector and the second video vector, a recommended video is obtained; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, the second video vector is a type vector corresponding to the type of video preferred by the user, the related video is a video related to the video watched by the user calculated using a collaborative filtering algorithm, and the similar video is a video having the same elements as the video watched by the user;

[0090] The identification information of the recommended video and a display tag corresponding to the recommended video are displayed on a display interface, wherein the display tag is generated according to the duration of the recommended video.

[0091] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0092] Clustering the first video vector and the second video vector, and determining whether the first video vector and the second video vector are of the same type according to the clustering result;

[0093] If the first video vector and the second video vector are of the same type, obtaining the recommended video from the related videos;

[0094] If the first video vector and the second video vector are not of the same type, obtaining the user's viewing time; and obtaining the recommended video according to the viewing time.

[0095] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0096] If the viewing time is less than a first time threshold, obtaining the recommended video from the similar videos and related videos of the user's historical viewing videos;

[0097] If the viewing time exceeds the first time threshold and is less than the second time threshold, obtaining the recommended video from the similar videos, wherein the second time threshold is greater than the first time threshold;

[0098] If the viewing time is greater than the second time threshold, the recommended video is obtained from the related videos.

[0099] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0100] Acquiring a user behavior, where the user behavior includes at least one of switching definition, dragging to fast forward, and multiple speed;

[0101] The recommended video is obtained from the related videos, the similar videos, the condensed short videos or the behind-the-scenes short videos according to the viewing time and the user behavior.

[0102] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0103] If the viewing time is less than or equal to a second time threshold, and the user behavior indicates that the user likes the video the user is watching, obtaining the recommended video from the related videos;

[0104] If the viewing time is less than or equal to the second time threshold, and the user behavior indicates that the user does not like the video the user is watching, obtaining the recommended video from the condensed short video;

[0105] If the viewing time is greater than the second time threshold, and the user behavior of switching definition is detected during the playback of the video watched by the user, obtaining the recommended video from the related videos and the original books related to the video watched by the user;

[0106] If the viewing time is greater than the second time threshold, and the user behavior of dragging to fast forward or double the speed is detected during the playback of the video watched by the user, or it is detected that the user watches until a preset time before the end of the video watched by the user, then the recommended video is obtained from the behind-the-scenes short video and the film reviews related to the video watched by the user.

[0107] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0108] If the viewing time is less than or equal to the second time threshold, determining a time corresponding to when the video viewed by the user is finished playing;

[0109] If the moment is within a preset time period, the recommended video is obtained from the condensed short video or the behind-the-scenes short video.

[0110] In an optional manner, the executable instruction causes the processor to perform the following operations:

[0111] Generate a display label for characterizing the duration type according to the duration type of the recommended video; and when detecting the user behavior of dragging the progress bar, display the identification information of the recommended video, the display label corresponding to the recommended video, and a direction arrow on the display interface playing the video watched by the user, wherein the direction arrow is used to indicate the direction of sliding from the dragging position on the progress bar to the identification information of the recommended video, and the display label is displayed on the identification information of the recommended video;

[0112] If a user behavior of sliding from the progress bar to the identification information of any of the recommended videos along the sliding direction of the direction arrow is detected, the player is controlled to play the recommended video.

[0113] The embodiment of the present invention obtains a first video vector while a user is watching a video; the first video vector is a type vector to which the video watched by the user belongs; a recommended video is obtained based on the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, and the second video vector is a type vector corresponding to the user's preferred video type; identification information of the recommended video and a display label corresponding to the recommended video are displayed on a display interface, the display label is generated based on the length of the recommended video, and can provide users with personalized related recommendation services based on the user's cognition of the video, the user's viewing environment and the user's behavior, thereby improving the user experience.

[0114] Figure 8 A schematic diagram of the structure of a computing device provided by an embodiment of the present invention is shown, and the specific embodiment of the present invention does not limit the specific implementation of the device.

[0115] like Figure 8 As shown, the computing device may include: a processor (processor) 802 , a communication interface (Communications Interface) 804 , a memory (memory) 806 , and a communication bus 808 .

[0116] The processor 802, the communication interface 804, and the memory 806 communicate with each other via a communication bus 808. The communication interface 804 is used to communicate with other devices such as a client or other server network elements. The processor 802 is used to execute a program 810, which can specifically execute the relevant steps in the above-mentioned video recommendation method embodiment.

[0117] Specifically, the program 810 may include program codes, which include computer operation instructions.

[0118] The processor 802 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. The one or more processors included in the device may be processors of the same type, such as one or more CPUs; or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0119] The memory 806 is used to store the program 810. The memory 806 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0120] The program 810 may be specifically configured to enable the processor 802 to perform the following operations:

[0121] In the process of a user watching a video, a first video vector is obtained; the first video vector is a type vector to which the video watched by the user belongs;

[0122] According to the first video vector and the second video vector, a recommended video is obtained; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, the second video vector is a type vector corresponding to the type of video preferred by the user, the related video is a video related to the video watched by the user calculated using a collaborative filtering algorithm, and the similar video is a video having the same elements as the video watched by the user;

[0123] The identification information of the recommended video and a display tag corresponding to the recommended video are displayed on a display interface, wherein the display tag is generated according to the duration of the recommended video.

[0124] In an optional manner, the program 810 enables the processor to perform the following operations:

[0125] Clustering the first video vector and the second video vector, and determining whether the first video vector and the second video vector are of the same type according to the clustering result;

[0126] If the first video vector and the second video vector are of the same type, obtaining the recommended video from the related videos;

[0127] If the first video vector and the second video vector are not of the same type, obtaining the user's viewing time; and obtaining the recommended video according to the viewing time.

[0128] In an optional manner, the program 810 enables the processor to perform the following operations:

[0129] If the viewing time is less than a first time threshold, obtaining the recommended video from the similar videos and related videos of the user's historical viewing videos;

[0130] If the viewing time exceeds the first time threshold and is less than the second time threshold, obtaining the recommended video from the similar videos, wherein the second time threshold is greater than the first time threshold;

[0131] If the viewing time is greater than the second time threshold, the recommended video is obtained from the related videos.

[0132] In an optional manner, the program 810 enables the processor to perform the following operations:

[0133] Acquiring a user behavior, where the user behavior includes at least one of switching definition, dragging to fast forward, and multiple speed;

[0134] The recommended video is obtained from the related videos, the similar videos, the condensed short videos or the behind-the-scenes short videos according to the viewing time and the user behavior.

[0135] In an optional manner, the program 810 enables the processor to perform the following operations:

[0136] If the viewing time is less than or equal to a second time threshold, and the user behavior indicates that the user likes the video the user is watching, obtaining the recommended video from the related videos;

[0137] If the viewing time is less than or equal to the second time threshold, and the user behavior indicates that the user does not like the video the user is watching, obtaining the recommended video from the condensed short video;

[0138] If the viewing time is greater than the second time threshold, and the user behavior of switching definition is detected during the playback of the video watched by the user, obtaining the recommended video from the related videos and the original books related to the video watched by the user;

[0139] If the viewing time is greater than the second time threshold, and the user behavior of dragging to fast forward or double the speed is detected during the playback of the video watched by the user, or it is detected that the user watches until a preset time before the end of the video watched by the user, then the recommended video is obtained from the behind-the-scenes short video and the film reviews related to the video watched by the user.

[0140] In an optional manner, the program 810 enables the processor to perform the following operations:

[0141] If the viewing time is less than or equal to the second time threshold, determining a time corresponding to when the video viewed by the user is finished playing;

[0142] If the moment is within a preset time period, the recommended video is obtained from the condensed short video or the behind-the-scenes short video.

[0143] In an optional manner, the program 810 enables the processor to perform the following operations:

[0144] Generate a display label for characterizing the duration type according to the duration type of the recommended video; and when detecting the user behavior of dragging the progress bar, display the identification information of the recommended video, the display label corresponding to the recommended video, and a direction arrow on the display interface playing the video watched by the user, wherein the direction arrow is used to indicate the direction of sliding from the dragging position on the progress bar to the identification information of the recommended video, and the display label is displayed on the identification information of the recommended video;

[0145] If a user behavior of sliding from the progress bar to the identification information of any of the recommended videos along the sliding direction of the direction arrow is detected, the player is controlled to play the recommended video.

[0146] The embodiment of the present invention obtains a first video vector while a user is watching a video; the first video vector is a type vector to which the video watched by the user belongs; a recommended video is obtained based on the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user, and the second video vector is a type vector corresponding to the user's preferred video type; identification information of the recommended video and a display label corresponding to the recommended video are displayed on a display interface, the display label is generated based on the length of the recommended video, and can provide users with personalized related recommendation services based on the user's cognition of the video, the user's viewing environment and the user's behavior, thereby improving the user experience.

[0147] The algorithm or display provided herein is not inherently related to any particular computer, virtual system or other equipment. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious to construct the structure required for this type of system. In addition, the embodiment of the present invention is not directed to any specific programming language yet. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description made to specific languages ​​above is for disclosing the best mode of the present invention.

[0148] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0149] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this method of disclosure should not be interpreted as reflecting an intention that the claimed invention requires more features than those expressly recited in each claim.

[0150] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0151] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

Claims

1. A video recommendation method, characterized in that: The method comprises: In the process of the user watching the video, a first video vector is obtained; the first video vector is a type vector of the video watched by the user. For the first video vector, the type of the video M1 watched by the user is obtained. The marking vector is a type and a number. The number is a type value obtained by quantifying the type of the video M1. The larger the number, the closer the video M1 is to the corresponding type. According to the first video vector and the second video vector, a recommended video is obtained; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user; the second video vector is a type vector corresponding to the type of video preferred by the user; for the second video vector, the type of video preferred by the user is obtained; the marking vector is a type and a number; the number is a preference type value obtained by quantifying the user preference; the larger the number, the stronger the user's preference for the type; the related video is a video related to the video watched by the user calculated using a collaborative filtering algorithm; the similar video is a video having the same elements as the video watched by the user; Displaying identification information of the recommended video and a display label corresponding to the recommended video on a display interface, wherein the display label is generated according to the duration of the recommended video; Obtaining a recommended video according to the first video vector and the second video vector, including: clustering the first video vector and the second video vector, and determining whether the first video vector and the second video vector are of the same type according to the clustering result; If the first video vector and the second video vector are of the same type, obtaining the recommended video from the related videos; If the first video vector and the second video vector are not of the same type, obtaining the user's viewing time; and obtaining the recommended video according to the viewing time.

2. The method according to claim 1, characterized in that The obtaining the recommended video according to the viewing time includes: If the viewing time is less than a first time threshold, obtaining the recommended video from the similar videos and related videos of the user's historical viewing videos; If the viewing time exceeds the first time threshold and is less than a second time threshold, obtaining the recommended video from the similar videos, wherein the second time threshold is greater than the first time threshold; If the viewing time is greater than the second time threshold, the recommended video is obtained from the related videos.

3. The method according to claim 2, characterized in that The step of obtaining the recommended video according to the viewing time also includes: Acquiring a user behavior, where the user behavior includes at least one of switching definition, dragging to fast forward, and multiple speed; The recommended video is obtained from the related videos, the similar videos, the condensed short videos or the behind-the-scenes short videos according to the viewing time and the user behavior.

4. The method according to claim 3, characterized in that The obtaining the recommended video from the related video, the similar video, the condensed short video or the behind-the-scenes short video according to the viewing time and the user behavior includes: If the viewing time is less than or equal to a second time threshold, and the user behavior indicates that the user likes the video the user is watching, obtaining the recommended video from the related videos; If the viewing time is less than or equal to the second time threshold, and the user behavior indicates that the user does not like the video the user is watching, obtaining the recommended video from the condensed short video; If the viewing time is greater than the second time threshold, and the user behavior of switching definition is detected during the playback of the video watched by the user, obtaining the recommended video from the related videos and the original books related to the video watched by the user; If the viewing time is greater than the second time threshold, and the user behavior of dragging to fast forward or double the speed is detected during the playback of the video watched by the user, or it is detected that the user watches until a preset time before the end of the video watched by the user, then the recommended video is obtained from the behind-the-scenes short video and the film reviews related to the video watched by the user.

5. The method according to claim 2, characterized in that: The method further comprises: If the viewing time is less than or equal to the second time threshold, determining a time corresponding to when the video viewed by the user is finished playing; If the moment is within a preset time period, the recommended video is obtained from the condensed short video or the behind-the-scenes short video.

6. The method according to claim 1, characterized in that The method further comprises: Generate a display label for characterizing the duration type according to the duration type of the recommended video; and when detecting the user behavior of dragging the progress bar, display the identification information of the recommended video, the display label corresponding to the recommended video, and a direction arrow on the display interface playing the video watched by the user, wherein the direction arrow is used to indicate the direction of sliding from the dragging position on the progress bar to the identification information of the recommended video, and the display label is displayed on the identification information of the recommended video; If a user behavior of sliding from the progress bar to the identification information of any of the recommended videos along the sliding direction of the direction arrow is detected, the player is controlled to play the recommended video.

7. A video recommendation device, characterized in that: The device comprises: A data acquisition unit is used to acquire a first video vector when a user is watching a video; the first video vector is a type vector of the video watched by the user, and for the first video vector, the type of the video M1 watched by the user is acquired, the marking vector is a type and a number, the number is a type value obtained by quantifying the type of the video M1, and the larger the number, the more similar the video M1 is to the corresponding type; A video recommendation unit is used to obtain a recommended video according to the first video vector and the second video vector; the recommended video includes at least one of a related video, a similar video, a condensed short video or a behind-the-scenes short video of the video watched by the user; the second video vector is a type vector corresponding to the user's preferred video type; for the second video vector, the user's preferred video type is obtained; the marking vector is a type and a number; the number is a preference type value obtained by quantifying the user's preference; the larger the number, the stronger the user's preference for the type; the related video is a video related to the video watched by the user calculated using a collaborative filtering algorithm; the similar video is a video having the same elements as the video watched by the user; A display unit, configured to display identification information of the recommended video and a display tag corresponding to the recommended video on a display interface, wherein the display tag is generated according to the duration of the recommended video; The video recommendation unit is also used to: cluster the first video vector and the second video vector, and determine whether the first video vector and the second video vector are of the same type based on the clustering result; if the first video vector and the second video vector are of the same type, obtain the recommended video from the related videos; if the first video vector and the second video vector are not of the same type, obtain the user's viewing time; and obtain the recommended video based on the viewing time.

8. A computing device, characterized in that include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the steps of the video recommendation method according to any one of claims 1-6.

9. A computer storage medium, characterized in that: The storage medium stores at least one executable instruction, and the executable instruction enables the processor to execute the steps of the video recommendation method according to any one of claims 1-6.

Images