阅读说明：本技术 一种视频信息推荐方法及系统 (Video information recommendation method and system ) 是由 姚娟娟 钟南山 于 2021-09-10 设计创作，主要内容包括：本发明提供一种视频信息推荐方法及系统,所述视频信息推荐方法包括：获取内容请求信息,所述内容请求信息包括用户请求的视频已播内容的时间节点和多个兴趣标签,其中,所述视频已播内容为已经完成播放的视频直播内容；根据所述时间节点和所述兴趣标签生成视频集锦内容,在视频直播内容之前播放所述视频集锦内容。将用户的兴趣标签和视频已播内容的内容标签进行匹配,获取适配用户兴趣标签的视频集锦内容,并在接入视频直播内容之前播放视频集锦内容,解决在点开直播之前了解视频直播内容的用户痛点,并通过兴趣标签与内容标签匹配的方法,提高了视频集锦内容的准确性。(The invention provides a video information recommendation method and a video information recommendation system, wherein the video information recommendation method comprises the following steps: acquiring content request information, wherein the content request information comprises a time node of video played content requested by a user and a plurality of interest tags, and the video played content is video live broadcast content which is played completely; and generating video collection content according to the time node and the interest tag, and playing the video collection content before the video live broadcast content. The interest tag of the user is matched with the content tag of the video broadcast content, the video collection content matched with the interest tag of the user is obtained, the video collection content is played before the video broadcast content is accessed, the problem that the user pain point of the video broadcast content is known before the video broadcast content is started is solved, and the accuracy of the video collection content is improved through the method of matching the interest tag with the content tag.)

1. A method for recommending video information, comprising:

acquiring content request information, wherein the content request information comprises a time node of video played content requested by a user and a plurality of interest tags, and the video played content is video live broadcast content which is played completely;

and generating video collection content according to the time node and the interest tag, and playing the video collection content before the video live broadcast content.

2. The video information recommendation method according to claim 1, wherein generating the video highlight content according to the time node and the interest tag comprises:

determining a plurality of content parts of the played content of the video according to the time nodes, splitting each content part into a plurality of content segments respectively, and acquiring content tags of the content segments;

determining a content segment according to the interest tag, wherein the content tag of the content segment corresponds to the interest tag;

and generating the video highlight content according to the content segment corresponding to the interest tag.

3. The method of claim 2, wherein determining the content segments according to the interest tags comprises:

obtaining the distance between the content tag and the interest tag, and determining similarity according to the distance;

and arranging the playing sequence of the plurality of content clips according to the similarity.

4. The video information recommendation method according to claim 3, further comprising:

when the user selects to skip the video highlight content, determining the content segment when the video highlight content is skipped over, and reducing the confidence coefficient of the interest tag;

and determining similarity according to the confidence degree and the distance.

5. The video information recommendation method according to claim 4, wherein said confidence level is mathematically expressed as:

wherein a2 is the confidence, t is the frequency of skipping playing by the current user, b is the average frequency of skipping playing by each user, and e is the natural logarithm;

the mathematical expression of the similarity is:

wherein a3 is the similarity, and a1 is the distance between the content tag and the interest tag.

6. The video information recommendation method of claim 3,

judging whether the duration of the video collection content is greater than a duration threshold value or not;

if so, removing the content segment with the lowest similarity according to the size of the similarity until the duration of the video collection content corresponds to the duration threshold;

and if not, increasing the content segment with the highest heat value until the time length of the video collection content corresponds to the time length threshold value, wherein the heat value comprises the number of online people or the number of bullet screens.

7. The video information recommendation method according to claim 3, wherein the distance between the content tag and the interest tag comprises: euclidean distance, manhattan distance, or checkerboard distance.

8. A video information recommendation system, comprising:

the system comprises a request module, a content request module and a content processing module, wherein the request module is used for acquiring content request information, the content request information comprises a time node of video broadcast content requested by a user and a plurality of interest tags, and the video broadcast content is video live broadcast content which is already played;

the generation module is used for generating video collection content according to the time node and the interest tag and playing the video collection content before video live broadcast content;

the request module is in signal connection with the generation module.

9. An electronic device, comprising:

one or more processors; and one or more machine readable media having instructions stored thereon that, when executed by the one or more processors, cause the electronic device to perform the video information recommendation method of any of claims 1-7.

10. A machine-readable medium having stored thereon instructions which, when executed by one or more processors, cause an apparatus to perform the video information recommendation method of any of claims 1-7.

Technical Field

The invention relates to the technical field of big data, in particular to a video information recommendation method and system.

Background

With the development of economic society and the progress of mobile internet and video technology, the method of taking forward service is also switched from online to online, for example, the taking forward service is realized by the methods of teaching, answering and the like through live video. However, due to the characteristic of the live video content, the user cannot clearly know whether the solicited service provided by the anchor broadcast matches the needs of the user at the first time, so that the user needs to know about the general view of the live video content before starting the live broadcast.

Disclosure of Invention

In view of the above drawbacks of the prior art, an object of the present invention is to provide a method and a system for recommending video information, which are used to solve the problem in the prior art that video information with a high degree of adaptation cannot be generated according to the interest of a user.

To achieve the above and other related objects, the present invention provides a video information recommendation method, including:

acquiring content request information, wherein the content request information comprises a time node of video played content requested by a user and a plurality of interest tags, and the video played content is video live broadcast content which is played completely;

and generating video collection content according to the time node and the interest tag, and playing the video collection content before the video live broadcast content.

Optionally, generating the video highlight content according to the time node and the interest tag includes:

determining a plurality of content parts of the played content of the video according to the time nodes, splitting each content part into a plurality of content segments respectively, and acquiring content tags of the content segments;

determining a content segment according to the interest tag, wherein the content tag of the content segment corresponds to the interest tag;

and generating the video highlight content according to the content segment corresponding to the interest tag.

Optionally, determining a content segment according to the interest tag includes:

obtaining the distance between the content tag and the interest tag, and determining similarity according to the distance;

and arranging the playing sequence of the plurality of content clips according to the similarity.

Optionally, the method further includes:

when the user selects to skip the video highlight content, determining the content segment when the video highlight content is skipped over, and reducing the confidence coefficient of the interest tag;

and determining similarity according to the confidence degree and the distance.

Optionally, the mathematical expression of the confidence coefficient is:

wherein a2 is the confidence, t is the frequency of skipping playing by the current user, b is the average frequency of skipping playing by each user, and e is the natural logarithm;

the mathematical expression of the similarity is:

wherein a3 is the similarity, and a1 is the distance between the content tag and the interest tag.

Optionally, judging whether the duration of the video collection content is greater than a duration threshold;

if so, removing the content segment with the lowest similarity according to the size of the similarity until the duration of the video collection content corresponds to the duration threshold;

and if not, increasing the content segment with the highest heat value until the duration of the video collection content corresponds to the duration threshold, wherein the heat value comprises the number of online people or the number of barrages.

Optionally, the distance between the content tag and the interest tag includes: euclidean distance, manhattan distance, or checkerboard distance.

A video information recommendation system comprising:

the system comprises a request module, a content request module and a content processing module, wherein the request module is used for acquiring content request information, the content request information comprises a time node of video broadcast content requested by a user and a plurality of interest tags, and the video broadcast content is video live broadcast content which is already played;

the generation module is used for generating video collection content according to the time node and the interest tag and playing the video collection content before video live broadcast content;

the request module is in signal connection with the generation module.

An electronic device, comprising:

one or more processors; and

one or more machine-readable media having instructions stored thereon that, when executed by the one or more processors, cause the electronic device to perform the video information recommendation method.

A machine-readable medium having stored thereon instructions, which when executed by one or more processors, cause an apparatus to perform the video information recommendation method.

As described above, the video information recommendation method and system of the present invention have the following beneficial effects:

the interest tag of the user is matched with the content tag of the video broadcast content, the video collection content matched with the interest tag of the user is obtained, the video collection content is played before the video broadcast content is accessed, the problem that the user pain point of the video broadcast content is known before the video broadcast content is started is solved, and the accuracy of the video collection content is improved through the method of matching the interest tag with the content tag.

Drawings

Fig. 1 is a schematic diagram illustrating a video information recommendation method according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of a video information recommendation system according to an embodiment of the present invention.

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention.

It should be noted that the drawings provided in the present embodiment are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated. The structures, proportions, sizes, and other dimensions shown in the drawings and described in the specification are for understanding and reading the present disclosure, and are not intended to limit the scope of the present disclosure, which is defined in the claims, and are not essential to the art, and any structural modifications, changes in proportions, or adjustments in size, which do not affect the efficacy and attainment of the same are intended to fall within the scope of the present disclosure. In addition, the terms "upper", "lower", "left", "right", "middle" and "one" used in the present specification are for clarity of description, and are not intended to limit the scope of the present invention, and the relative relationship between the terms and the terms is not to be construed as a scope of the present invention.

Referring to fig. 1, the present invention provides a video information recommendation method, including:

s1: acquiring content request information, wherein the content request information comprises a time node of video played content requested by a user and a plurality of interest tags, and the video played content is video live broadcast content which is played completely;

s2: and generating video collection content according to the time node and the interest tag, and playing the video collection content before the video live broadcast content.

For example, in the process of logging in the live broadcast room by the user, the system acquires the time for accessing the live broadcast room as the time node of the video played content requested by the user, and acquires a plurality of interest tags of the user, during the login process, the content request information is sent to the server, in order to display the video played content of the video live content, the video collection content can be obtained from the video played content which is just finished to be played, in order to meet the accuracy of the video collection content, the interest label of the user is matched with the content label of the video played content to obtain the video collection content which is adapted to the interest label of the user, and plays the video collection content before accessing the video live broadcast content, thereby solving the user pain of knowing the video live broadcast content before starting the live broadcast, and the accuracy of the video collection content is improved by the method of matching the interest tag with the content tag.

In some implementations, generating the video highlight content according to the time node and the interest tag includes:

determining a plurality of content parts of the played content of the video according to the time nodes, splitting each content part into a plurality of content segments respectively, and acquiring content tags of the content segments; because the live video broadcast has obvious time step and hot value step, for example, a plurality of content parts can be divided by the hot value wave crest and wave trough of the live video broadcast, and each content part is divided into a plurality of content segments, and a time period of the live video broadcast can be set, and the content parts are divided by the time period, and each content part is divided into a plurality of content segments;

determining content segments according to the interest tags, wherein the content tags of the content segments correspond to the interest tags, and the content tags of each content segment can be obtained by performing target identification on video frames of the content segments or obtaining letter captions or bullet screen information in a natural language processing mode to complete automatic marking of the tags, wherein the interest tags can be obtained in a mode that: obtaining interest tags of the user from user historical data, or filling related tags by the user in a self-describing manner;

and generating the video highlight content according to the content segment corresponding to the interest tag.

In some implementations, determining a content segment from the interest tag includes:

obtaining the distance between the content tag and the interest tag, and determining the similarity according to the distance, for example, the content tag and the interest tag may be vectorized, and the distance between the two vectors is calculated, where the closer the distance is, the greater the similarity between the two vectors is, and the calculation manner of the distance between the vectors may be performed by one of the following manners: euclidean distance, manhattan distance, or chessboard distance;

and arranging the playing sequence of the plurality of content clips according to the similarity, acquiring the content clips of the content tags with high similarity to the interest tags, and playing the content clips at the front time sequence in the video collection, so that the pain points and the actual requirements of the user can be grasped.

In some implementations, due to the limitation of the interest tag, the recommended content of the video highlights is not accurate enough, and the user chooses to skip or opt out of the live broadcast room when watching the video highlights, and in order to further adapt to the user's needs, the video information recommendation method further includes:

when the user selects to skip the video highlight content, determining the content segment when the video highlight content is skipped over, and reducing the confidence coefficient of the interest tag;

and determining similarity according to the confidence degree and the distance. By means of setting confidence, actual requirements of users are further determined, and the problem that generated video collection is inaccurate due to inaccurate interest tags is avoided.

In some implementations, the mathematical expression of the confidence level is:

wherein a2 is the confidence, t is the frequency of skipping playing by the current user, b is the average frequency of skipping playing by each user, and e is the natural logarithm;

the mathematical expression of the similarity is:

wherein a3 is the similarity, and a1 is the distance between the content tag and the interest tag. For example, if the frequency of skipping playing by a user is 1, and the average frequency of skipping playing by each user is 2, it indicates that the recommended video highlight for the user is relatively accurate, and if it is necessary to raise the confidence for the interest tag, t =1, b =2, a2 > a1, a3 > a1 raises the occurrence timing of the content segment;

for another example, if the frequency of skipping playing by a user is 2 and the average frequency of skipping playing by each user is 1, it indicates that the video highlights recommended for the user are relatively inaccurate, and if it is necessary to reduce the confidence level for the interest tag, t =2, b =1, a1 > a2, a1 > a3 reduces the occurrence timing of the content segment.

In some implementation processes, in order to ensure the normativity of the video highlight segments, generally, the acceptance degree of a short video user of 14 seconds is considered to be the highest, and whether the duration of the video highlight content is greater than a duration threshold value or not can be judged;

if so, removing the content segment with the lowest similarity according to the size of the similarity until the duration of the video collection content corresponds to the duration threshold;

and if not, increasing the content segment with the highest heat value until the duration of the video collection content corresponds to the duration threshold, wherein the heat value comprises the number of online people or the number of barrages. And selecting some content segments meeting the similarity requirement through the similarity to generate the video collection meeting the time length threshold, and on the other hand, making up the generation requirement when the time length of the video collection content is insufficient through the heat value. The standardization of video collection information is improved, and the requirements of customers are met.

Referring to fig. 2, the present invention provides a video information recommendation system, including:

the system comprises a request module, a content request module and a content processing module, wherein the request module is used for acquiring content request information, the content request information comprises a time node of video broadcast content requested by a user and a plurality of interest tags, and the video broadcast content is video live broadcast content which is already played;

the generation module is used for generating video collection content according to the time node and the interest tag and playing the video collection content before video live broadcast content;

the request module is in signal connection with the generation module.

Optionally, generating the video highlight content according to the time node and the interest tag includes:

determining a plurality of content parts of the played content of the video according to the time nodes, splitting each content part into a plurality of content segments respectively, and acquiring content tags of the content segments;

determining a content segment according to the interest tag, wherein the content tag of the content segment corresponds to the interest tag;

and generating the video highlight content according to the content segment corresponding to the interest tag.

Optionally, determining a content segment according to the interest tag includes:

obtaining the distance between the content tag and the interest tag, and determining similarity according to the distance;

and arranging the playing sequence of the plurality of content clips according to the similarity.

Optionally, the method further includes:

when the user selects to skip the video highlight content, determining the content segment when the video highlight content is skipped over, and reducing the confidence coefficient of the interest tag;

and determining similarity according to the confidence degree and the distance.

Optionally, the mathematical expression of the confidence coefficient is:

wherein a2 is the confidence, t is the frequency of skipping playing by the current user, b is the average frequency of skipping playing by each user, and e is the natural logarithm;

the mathematical expression of the similarity is:

wherein a3 is the similarity, and a1 is the distance between the content tag and the interest tag.

Optionally, judging whether the duration of the video collection content is greater than a duration threshold;

if so, removing the content segment with the lowest similarity according to the size of the similarity until the duration of the video collection content corresponds to the duration threshold;

and if not, increasing the content segment with the highest heat value until the duration of the video collection content corresponds to the duration threshold, wherein the heat value comprises the number of online people or the number of barrages.

Optionally, the distance between the content tag and the interest tag includes: euclidean distance, manhattan distance, or checkerboard distance.

An embodiment of the present invention provides an electronic device, including: one or more processors; and one or more machine readable media having instructions stored thereon that, when executed by the one or more processors, cause the electronic device to perform one or more of the methods. The invention is operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

Embodiments of the invention also provide one or more machine-readable media having instructions stored thereon, which when executed by one or more processors, cause an apparatus to perform one or more of the methods described herein. The invention may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.