WO2022228390A1

WO2022228390A1 - Media content processing method, apparatus and device, and storage medium

Info

Publication number: WO2022228390A1
Application number: PCT/CN2022/088995
Authority: WO
Inventors: 朱亚光; 李孟杰; 严冰; 黄胜兰; 李小成
Original assignee: 北京字跳网络技术有限公司
Priority date: 2021-04-26
Filing date: 2022-04-25
Publication date: 2022-11-03
Also published as: CN115250295A

Abstract

Embodiments of the present disclosure provide a media content processing method, apparatus and device, and a storage medium. The media content processing method comprises: determining a popularity level of target media content; on the basis of the popularity level of the target media content, determining a target CDN node type corresponding to the target media content, wherein popularity levels have one-to-one correspondence to CDN node types; on the basis of the target CDN node type, determining a target CDN node corresponding to the target media content, and sending node information of the target CDN node to a client, so that the client obtain the target media content from the target CDN node on the basis of the node information.

Description

Media content processing method, apparatus, device and storage medium

This application claims the priority of the Chinese Patent Application No. 202110454466.9 filed with the China Patent Office on April 26, 2021, the entire contents of which are incorporated herein by reference.

technical field

The embodiments of the present disclosure relate to the technical field of streaming media processing, for example, to a media content processing method, apparatus, device, and storage medium.

Background technique

With the rapid development of the Internet, the bandwidth occupied by streaming media applications has exploded. In order to transmit media content data to clients as quickly as possible, streaming media applications generally use Content Delivery Network (CDN) technology.

SUMMARY OF THE INVENTION

Embodiments of the present disclosure provide a media content processing method, apparatus, device, and storage medium, which improve the hit rate of media content in a CDN node and reduce back-to-source bandwidth.

In a first aspect, an embodiment of the present disclosure provides a media content processing method, including:

Determine the popularity level of the target media content;

Determine the target content distribution network CDN node type corresponding to the target media content based on the popularity level of the target media content, wherein the popularity level corresponds to the CDN node type one-to-one;

Determine the target CDN node corresponding to the target media content based on the target CDN node type, and send the node information of the target CDN node to the client, so that the client can access the target CDN based on the node information The target media content is obtained from the node.

In a second aspect, an embodiment of the present disclosure further provides a media content processing apparatus, including:

The heat level determination module is set to determine the heat level of the target media content;

A node type determination module, configured to determine the target content distribution network CDN node type corresponding to the target media content based on the popularity level of the target media content, wherein the popularity level corresponds to the CDN node type one-to-one;

A target node determination module, configured to determine the target CDN node corresponding to the target media content based on the target CDN node type, and send the node information of the target CDN node to the client, so that the client is based on the target CDN node. The node information acquires the target media content from the target CDN node.

In a third aspect, an embodiment of the present disclosure further provides a media content processing device, including:

one or more processors;

memory, arranged to store one or more programs;

When the one or more programs are executed by the one or more processors, the one or more processors implement the media content processing method according to any one of the embodiments of the present disclosure.

In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any one of the embodiments of the present disclosure is implemented The media content processing method.

Description of drawings

Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that the originals and elements are not necessarily drawn to scale.

Fig. 1 is a CDN typical topology diagram provided by an embodiment of the present disclosure;

FIG. 2 is a flowchart of a media content processing method provided by an embodiment of the present disclosure;

3 is a flowchart of another media content processing method provided by an embodiment of the present disclosure;

4a is a schematic diagram of a CDN node in the prior art;

FIG. 4b is a schematic diagram of a CDN node classification provided by an embodiment of the present disclosure;

5 is a structural diagram of a media content processing device provided by an embodiment of the present disclosure;

FIG. 6 is a structural diagram of a media content processing device provided by an embodiment of the present disclosure.

Detailed ways

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for the purpose of A more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes, and are not intended to limit the protection scope of the present disclosure.

It should be understood that the multiple steps described in the method embodiments of the present disclosure may be performed in different orders and/or in parallel. Furthermore, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this regard.

As used herein, the term "including" and variations thereof are open-ended inclusions, ie, "including but not limited to". The term "based on" is "based at least in part on." The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions of other terms will be given in the description below.

It should be noted that concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or interdependence.

It should be noted that the modifications of "a" and "a plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless the context clearly indicates otherwise, they should be understood as "one or a plurality of". multiple".

The names of messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes, and are not intended to limit the scope of these messages or information.

The CDN technology in the related art caches the recently accessed media content on the CDN node, so that when the client obtains the media content, it can directly obtain the media content from the CDN node, without the need to return data to the source, thereby reducing the impact on the source site. number of visits.

The CDN technology in the related art caches all media content in the CDN node indiscriminately. When new media content enters the CDN node, the media content with low heat level may be squeezed out of the CDN node, resulting in low heat level. The media content is not on the CDN node. When the client accesses media content with a low popularity level, the media content cannot be found on the CDN node, resulting in a low hit rate of the media content with a low popularity level. Increase back-to-source bandwidth.

First, the application scenarios of the embodiments of the present disclosure are briefly explained. The embodiments of the present disclosure can be applied to any scenario of accessing a content distribution network to obtain media content, for example, it can be applied to a feed streaming scenario, can also be applied to a scenario of obtaining media content through a media playback platform or a website, and can also be applied to Click on the media content scene from the home page of the media content publisher.

The “feed” in the embodiment of the present disclosure may be a content aggregator that combines several news sources actively subscribed by the user to help the user continuously obtain the latest feed content. The feed is a Really Simple Syndication (RSS) The interface used to receive the source of this information.

The feed stream in the embodiment of the present disclosure, also referred to as an information stream, is a continuously updated information stream, which can push the information in the RSS to the user.

In the feed stream scenario, the server's recommendation system pushes a series of media content (such as videos) to the client. When the user watches the media content in the feed stream, the client responds to the user's operation and sends a series of media content to the server. For a media content request, the server determines the popularity level of the media content corresponding to the media content request in response to the media content request, obtains the media content from a corresponding type of CDN node according to the popularity level, and distributes it to the client.

FIG. 1 is a typical topology structure diagram of a CDN provided by an embodiment of the present disclosure. As shown in Figure 1, the CDN technology principle is: when a user requests a certain Uniform Resource Locator (URL), a media content request is initiated to the CDN node, and the CDN node will detect the corresponding media content request sent by the user. Whether the media content is expired, if not, it directly responds to the media content request, and returns the media content corresponding to the media content request to the client. At this point, a complete http request ends. If the media content corresponding to the media content request has expired, the CDN node needs to send a back to the source request to the origin site to pull the media content corresponding to the latest media content request, update the locally cached media content, and send The media content corresponding to the latest media content request is returned to the client. The expiration of the requested content means that the media content corresponding to the media content request cannot be queried at the CDN node.

With the above CDN technology, all media content pulled by the origin site is indiscriminately cached in the CDN node. When new media content enters the CDN node, the media content with a low level of popularity may be squeezed out of the CDN node, resulting in a low level of popularity. The media content is not on the CDN node. When the client requests media content with a low popularity level, the media content cannot be found on the CDN node, resulting in a low hit rate of the media content with a low popularity level. Increase back-to-source bandwidth.

Embodiments of the present disclosure provide a media content processing method, apparatus, device, and storage medium. The embodiments of the present disclosure differentiate media content by degree of popularity, based on the corresponding relationship between the degree of popularity and node type, in different types of CDNs. Searching for the media content corresponding to the popularity in the node can improve the hit rate of the media content in the CDN node and reduce the back-to-source bandwidth.

The media content processing method, apparatus, device, and storage medium provided by the embodiments of the present disclosure will be described in detail below with reference to specific embodiments.

FIG. 2 is a flowchart of a method for processing media content provided by an embodiment of the present disclosure. This embodiment is applicable to a situation in which a client requests media content from a CDN node. The method may be executed by a media content processing apparatus. The media content processing apparatus may be implemented by means of software and/or hardware. The media content processing method is applied in a server.

As shown in FIG. 2 , the media content processing method provided by this embodiment mainly includes steps S11 , S12 , S13 and S14 .

S11. Determine the popularity level of the target media content.

In this embodiment, the media content refers to the content displayed on the client, which may be one or a combination of video, audio, text, and cards. It should be noted that the media content in this embodiment may be the media content in the feed stream, or may be the media content of a certain video website or the homepage of a video publisher. The target media content refers to the media content requested by the client from the server.

In one embodiment, a method of determining target media content is provided. The server receives the media content request sent by the client, parses the media content request, and determines the media content corresponding to the media content request as the target media content.

In another embodiment, a method of determining target media content is provided. The feed stream request sent by the client is received, the feed stream request is parsed, each media content in the feed stream is determined as the target media content, and steps S12, S13 and S14 are performed in sequence.

Among them, the popularity level is used to represent the attention degree of each media content. The heat level can be divided into multiple levels according to the needs, for example: it can be divided into four levels: S, A, B, and F. Among them, the S level is the highest, and the A level is lower than the S level. The B grade is lower than the A grade, and the F grade is the lowest grade. The higher the level, the more times the media content is viewed, and the lower the level, the less the media content is viewed. It can also be divided into three levels: high, medium and low. This embodiment only describes the level division, but is not limited.

In this embodiment, for example, the heat level is divided into two levels, hot and cold. Hot media content refers to media content with a higher degree of attention in a short period of time. For example, hot media content can be media content with tens of thousands of playbacks within an hour, and hot media content can also be commented and clicked within an hour. Media content with a significant increase in likes. Cold media content refers to media content that has received less attention in a short period of time. For example, cold media content can be media content with very low playback volume within an hour, and cold media content can also be media content with comment volume within an hour, and the growth rate of likes is very slow or does not increase.

In one embodiment, through the relevant features of the target media content, a gradient boosting decision tree (Gradient Boosting Decision Tree, GBDT) tree model is used to predict the growth amount of the media content playback in the next hour, based on the correspondence between the growth amount and the popularity level relationship to determine the popularity level corresponding to the target media content.

For example, it may be judged whether the play increase amount exceeds a preset play increase amount threshold. In response to determining that the playback increase exceeds the preset playback increase threshold, the media content has a heat level of hot media content, and in response to determining that the playback increase is lower than the preset playback increase threshold, the media content has a heat level of cold. media content. The way of judging the popularity level according to the increase in comments and the increase in likes is similar to the above playback increment, and will not be repeated here.

For example, multiple growth amount ranges can also be set to divide the popularity level of the media content in more detail. For example, four growth ranges are set, and the media content can be divided into four levels: S, A, B, and F.

In another embodiment, by normalizing the relevant features of the target media content, and assigning different weights to each relevant feature, and performing weighting processing based on the normalized relevant features and their corresponding weights, Obtain the heat index corresponding to the target media content. Based on the correspondence between the popularity index and the popularity level, the popularity level corresponding to the target media content is determined.

It should be noted that the division method of the corresponding relationship between the heat index and the heat level is basically the same as that of the above-mentioned corresponding relationship between the increase amount and the heat level. For details, please refer to the above description, which will not be repeated in this implementation.

S12. Determine a target content distribution network CDN node type corresponding to the target media content based on the popularity level of the target media content, where the popularity level corresponds to the CDN node type one-to-one.

Among them, all CDN nodes in the CDN system are classified, and there is a corresponding relationship between the CDN node type and the heat level. For example, there is a one-to-one correspondence between popularity levels and CDN node types. For example, a CDN node group includes at least one CDN node.

For example, one popularity level corresponds to one CDN node type. For example, the heat levels S, A, B, and F correspond to the first type of CDN node, the second type of CDN node, the third type of CDN node, and the fourth type of CDN node of the CDN node type, respectively. Another example: hot media content corresponds to a hot type CDN node, and cold media content corresponds to a cold type CDN node.

For example, the corresponding relationship between the popularity level and the CDN node type is preselected and stored in the server. After determining the popularity level of the target media content, the server searches for the target CDN node type corresponding to the target media content in the correspondence between the popularity level and the CDN node type.

S13. Determine the target CDN node corresponding to the target media content based on the target CDN node type, and send the node information of the target CDN node to the client, so that the client can download from the target based on the node information The target media content is acquired from the target CDN node.

In this embodiment, the target CDN node may be understood as the CDN node included in the target CDN node type, that is, the type to which the target CDN node belongs is the target CDN node type.

For example, the hot type CDN node includes the first CDN node, the second CDN node, the third CDN node, and the fourth CDN node, and the cold type CDN node includes the fifth CDN node and the sixth CDN node. It should be noted that a CDN node type includes one or more CDN nodes.

Determining the target CDN node corresponding to the target media content based on the target CDN node type may be to determine all CDN nodes included in the target CDN node type as target CDN nodes corresponding to the target media content.

After the target CDN node corresponding to the target media content is determined, node information (eg, node identification information) of the target CDN node may be sent to the client. Therefore, the client can request the corresponding target CDN node to acquire the target media content according to the node information.

In one embodiment, when the server determines that the popularity level of the target media content is the hot media content, it determines the hot type CDN node corresponding to the target media content, then determines the CDN node included in the hot type CDN node as the target CDN node, and sets The node information of the target CDN node is sent to the client. Therefore, the client can send a request for acquiring the target media content to the target CDN node based on the node information. Correspondingly, when the target CDN node receives the request from the client to obtain the target media content, it can search for the target media content, and if the target media content is found, transmit the target media content to the client; if the target media content is not found, The target CDN node sends a back to the source request to the origin site to pull the latest target media content, cache the latest target media content, and at the same time, return the latest target media content to the client.

For example, when the server determines that the target media content is cold media content, it determines that the target media content is cached in the cold type CDN node, determines the CDN node included in the cold type CDN node as the target CDN node, and sends the node information of the target CDN node. to the client. Therefore, the client can send a request for acquiring the target media content to the target CDN node based on the node information. Correspondingly, when the target CDN node receives the request from the client to obtain the target media content, it can search for the target media content, and if the target media content is found, transmit the target media content to the client; if the target media content is not found, The target CDN node sends a back-to-origin request to the origin site to pull the latest target media content, cache the latest target media content, and at the same time, return the latest target media content to the client.

On the basis of the above-mentioned embodiments, another scheduling manner of media content is provided. When the server determines that the target media content is hot media content, it determines that the target media content is cached in the CDN nodes included in the hot type CDN node, then determines the CDN nodes included in the hot type CDN node as the target CDN node, and the target CDN node node information is sent to the client. Therefore, the client can send a request for acquiring the target media content to the CDN nodes included in the hot type CDN node based on the node information. Correspondingly, when the CDN node included in the hot type CDN node receives the request from the client to obtain the target media content, it searches for the target media content. If the target media content is found, the target media content is transmitted to the client; if the target media content is not found. For the target media content, a back-to-origin request is sent to the origin site to pull the latest target media content, cache the latest target media content, and at the same time, return the latest target media content to the client.

For example, when the server determines that the target media content is cold media content, and determines that the target media content is cached in the CDN nodes included in the cold-type CDN node, the CDN nodes included in the cold-type CDN node are determined as the target CDN node, and the target CDN node's Node information is sent to the client. Therefore, the client can send a request for acquiring the target media content to the CDN nodes included in the cold-type CDN node based on the node information. Correspondingly, the CDN node included in the cold type CDN node searches for the target media content when receiving the request from the client to obtain the target media content, and if the target media content is found, transmits the target media content to the client; For the target media content, a back-to-origin request is sent to the origin site to pull the latest target media content, cache the latest target media content, and at the same time, return the latest target media content to the client.

An embodiment of the present disclosure provides a media content processing method, including: determining a popularity level of a target media content; determining a target CDN node type corresponding to the target media content based on the popularity level of the target media content, wherein the popularity level and the CDN node type are one-to-one Corresponding; determine the target CDN node corresponding to the target media content based on the target CDN node type, and send the node information of the target CDN node to the client, so that the client obtains the target media content from the target CDN node based on the node information. The embodiment of the present disclosure can improve the hit rate of the media content in CDN nodes by distinguishing the media content by the popularity level, and searching for the media content corresponding to the popularity in different types of CDN nodes based on the corresponding relationship between the popularity level and the node type. , reducing the back-to-source bandwidth.

On the basis of the above-mentioned embodiment, the embodiment of the present disclosure further refines the media content processing method. , using a pre-trained tree model to predict the popularity level of the target media content, wherein the media content features include one or more of the following: media content visits, media content viewing, media content likes, media content comments volume, media content downloads, media content sharing”.

When training the tree model, the media content feature and the real media content popularity level identifier can be used as input to obtain the media content predicted popularity level output, based on the loss function, the real media content popularity level and the predicted media content popularity level Adjust the model parameters until the model output is as expected, and get the trained tree model. After the model is trained, the model can be used to predict the popularity level of the media content, and the media content feature is used as an input to obtain the popularity level of the model. For example, during training and use, the tree model predicts the hotness or coldness of the media content by predicting whether the playback volume of the media content in a preset time period in the future exceeds a playback volume threshold.

In this embodiment, the media content access volume may be the number of times the client accesses the media content, and the access volume may be determined by counting the number of times the client accesses the CDN address corresponding to the media content, and the specific statistical method is not performed in this embodiment. limited. The media content viewing volume refers to the number of times the client plays the media content.

Among them, the tree model is also called the decision tree model, and the decision tree model is a simple and easy-to-use non-parametric classifier. In this embodiment, any decision tree model can be selected for training to obtain a specific training method of the pre-trained tree model, which is not limited in this embodiment.

In one embodiment, the amount of media content visits, the amount of media content viewing, the amount of media content likes, the amount of media content comments, the amount of media content downloads, and the slope of the amount of media content sharing are used as media content features, and are input into the pre-trained tree. Model, the pre-trained tree model directly outputs the popularity level of the target media content.

In one embodiment, the amount of media content visits, the amount of media content viewing, the amount of media content likes, the amount of media content comments, the amount of media content downloads, and the slope of the amount of media content sharing are used as media content features, and are input into the pre-trained tree. Model, the pre-trained tree model outputs the target media content playback growth amount, and determines the popularity level of the target media content based on the playback growth amount.

In the embodiment of the present disclosure, the pre-trained tree model is used to predict the popularity level of the target media content, which can improve the accuracy of the prediction required for the hot and cold prediction of the target media content.

FIG. 3 is a flowchart of another media content processing method provided by an embodiment of the present disclosure. As shown in FIG. 3 , another media content processing method provided by an embodiment of the present disclosure mainly includes the following steps:

S21. Use the pre-trained tree model to predict the relationship between the play increase amount of the target media content and the play increase amount threshold within a preset time period in the future.

In this embodiment, the preset time period in the future may refer to one hour in the future, or two hours in the future. In this embodiment, it can be within an hour in the future. The play increment may refer to the play increment of the media content within a preset time period in the future, for example, the play increment of the target media content in the next hour. The relationship between the playback increase and the playback increase threshold includes: the playback increase exceeds the playback increase threshold, or the playback increase is lower than the playback increase threshold.

In one embodiment, the amount of media content visits, the amount of media content viewing, the amount of media content likes, the amount of media content comments, the amount of media content downloads, and the slope of the amount of media content sharing are used as media content features, and are input into the pre-trained tree. Model, the pre-trained tree model outputs the relationship between the playback growth of the target media content and the playback growth threshold.

S22. Determine the popularity level of the target media content based on the relationship between the play increase amount of the target media content and the play increase amount threshold.

In one embodiment, determining the popularity level of the target media content based on the relationship between the play increase amount of the target media content and the play increase amount threshold includes: in response to determining that the play increase amount of the target media content is greater than or equal to An increase amount threshold, it is determined that the popularity level of the target media content is the first popularity; in response to determining that the playback increase amount of the target media content is less than the increase amount threshold, it is determined that the popularity level of the target media content is the second popularity.

For example, it can be determined whether the increase amount exceeds the increase amount threshold, and in response to determining that the increase amount exceeds the increase amount threshold, the popularity level of the media content is the first popularity level, that is, the media content is hot media content, in response to determining that the increase amount is lower than the increase amount threshold. The increase amount threshold, the popularity level of the media content is the second popularity, that is, the media content is cold media content.

Fig. 4a is a schematic diagram of a CDN node in the prior art, and Fig. 4b is a schematic diagram of a CDN node classification provided by an embodiment of the present disclosure; as shown in Fig. 4a, each original CDN node in the CDN system contains hot media content and cold media content. After classifying the CDN nodes, as shown in Figure 4b, the CDN nodes are divided into hot-type CDN nodes that cache hot media content and cold-type CDN nodes that cache cold media content. When the target media content is predicted to be hot media content, it is obtained from the hot type CDN node, and the target media content is predicted to be cold media content and obtained from the cold type CDN node, so that the resources are re-divided. Bringing the cold media content back to one node prevents the mixed cold and hot nodes from squeezing the cold resources out of the CDN's cache, resulting in a decrease in the hit rate of the cold resource CDN.

S23. Determine a target content distribution network CDN node type corresponding to the target media content based on the popularity level of the target media content; wherein the popularity level corresponds to the CDN node type one-to-one.

S24. Determine the target CDN node corresponding to the target media content based on the target CDN node type, and send the node information of the target CDN node to the client, so that the client can download from the target based on the node information The target media content is acquired from the target CDN node.

On the basis of the above embodiment, the method further includes: a training process of the tree model. The tree model training method provided by the embodiments of the present disclosure mainly includes: using a binary cross-entropy loss function to train the tree model.

In this embodiment, the tree model can be trained by using the tree model training samples constructed from multiple media contents and the binary cross-entropy loss function determined based on the weight of each media content. At this time, for example, using a binary cross-entropy loss function to train the tree model to obtain a pre-trained tree model, including: constructing a tree model training sample by using multiple media contents; The feature determines the weight of each media content, and determines the binary cross-entropy loss function according to the weight of each media content; the tree model is trained based on the tree model training samples and the binary-class cross entropy loss function to obtain Pretrained tree model.

In one embodiment, the binary cross-entropy loss function is:

Among them, N is the total amount of media content in the tree model training sample, _yi represents the label of the ith media content in the tree model training sample, and pi represents the probability that the _ith media content in the tree model training sample is predicted to be a positive example , α _i represents the weight of the i-th media content in the tree model training sample,

vv represents the viewing volume of the ith media content sample in the tree model training sample in a preset time period in the future, VV represents the total viewing volume of the ith media content sample in the tree model training sample, and T represents the ith media content sample in the tree model training sample. The total duration corresponding to the total viewing amount of media content samples, t represents the viewing duration of the i-th media content sample in the tree model training sample from its creation to the current time point.

It should be noted that this embodiment can also use the existing GBDT tree model, which will not be repeated here.

In this embodiment, more weights are set for the loss of media content with a higher proportion of the viewing volume vv in the future preset time period, thereby improving the prediction effect of the later media content with a relatively high viewing volume.

FIG. 5 is a structural diagram of a media content processing apparatus provided by an embodiment of the present disclosure. This embodiment is applicable to a situation where a client requests media content from a CDN node, and the media content processing apparatus can be processed through software and/or implemented in hardware. The media content processing device is integrated in the server.

As shown in FIG. 5 , the media content processing apparatus provided in this embodiment mainly includes a heat level determination module 51 , a node type determination module 52 and a target node determination module 53 .

Wherein, the heat level determination module 51 is set to determine the heat level of the target media content;

The node type determination module 52 is configured to determine the target content distribution network CDN node type corresponding to the target media content based on the popularity level of the target media content, wherein the popularity level corresponds to the CDN node type one-to-one;

The target node determination module 53 is configured to determine the target CDN node corresponding to the target media content based on the target CDN node type, and send the node information of the target CDN node to the client, so that the client is based on the target CDN node. The node information acquires the target media content from the target CDN node.

An embodiment of the present disclosure provides a media content processing apparatus, which is mainly configured to perform the following operations: determine the popularity level of the target media content; determine the target CDN node type corresponding to the target media content based on the popularity level of the target media content, wherein the popularity level is the same as One-to-one correspondence between the types of CDN nodes; determine the target CDN node corresponding to the target media content based on the target CDN node type, and send the node information of the target CDN node to the client, so that the client can select the target CDN node based on the node information. obtain the target media content from the target CDN node. The embodiment of the present disclosure can improve the hit rate of the media content in CDN nodes by distinguishing the media content by the popularity level, and searching for the media content corresponding to the popularity in different types of CDN nodes based on the corresponding relationship between the popularity level and the node type. , reducing the back-to-source bandwidth.

In one embodiment, the popularity level determination module 51 is configured to use a pre-trained tree model to predict the popularity level of the target media content based on the media content characteristics of the target media content, wherein the media content characteristics include the following One or more of: media content visits, media content viewings, media content likes, media content comments, media content downloads, and media content sharing.

In one embodiment, the heat level determination module 51 includes:

an increase amount prediction unit, configured to predict the relationship between the play increase amount and the play increase amount threshold of the target media content in a preset time period in the future by using the pre-trained tree model;

The popularity level determination unit is configured to determine the popularity level of the target media content based on the relationship between the play increase amount of the target media content and the play increase amount threshold.

For example, the popularity level determination unit is configured to determine that the popularity level of the target media content is the first popularity in response to determining that the play increase of the target media content is greater than or equal to an increase threshold; The play growth amount is less than the growth amount threshold, and it is determined that the popularity level of the target media content is the second popularity.

In one embodiment, the apparatus further includes: a model training module, configured as:

The tree model is trained using a binary cross-entropy loss function to obtain a pre-trained tree model.

In one embodiment, the model training module is configured to: construct a tree model training sample by using a plurality of media contents; determine the weight of each media content according to the media content characteristics of the plurality of media contents, and The weight of the content determines a binary cross-entropy loss function; the tree model is trained based on the tree model training sample and the binary cross-entropy loss function to obtain a pre-trained tree model.

In one embodiment, the binary cross-entropy loss function is:

The media content processing apparatus provided in this embodiment can execute the media content processing method provided by any embodiment of the present disclosure, and has corresponding functional modules and beneficial effects for executing the media content processing method.

Referring next to FIG. 6 , it shows a schematic structural diagram of an electronic device (eg, a terminal device or a server in FIG. 6 ) 600 suitable for implementing an embodiment of the present disclosure. Terminal devices in the embodiments of the present disclosure may include, but are not limited to, such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablets), PMPs (portable multimedia players), vehicle-mounted terminals (eg, mobile terminals such as in-vehicle navigation terminals), etc., and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in FIG. 6 is only an example, and should not impose any limitation on the function and scope of use of the embodiments of the present disclosure.

As shown in FIG. 6, an electronic device 600 may include a processing device (eg, a central processing unit, a graphics processor, etc.) 601 that may be loaded into random access according to a program stored in a read only memory (ROM) 602 or from a storage device 608 A program in a memory (RAM) 603 executes various appropriate actions and processes. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input/output (I/O) interface 605 is also connected to bus 604 .

Typically, the following devices can be connected to the I/O interface 605: input devices 606 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; including, for example, a liquid crystal display (LCD), speakers, vibration An output device 607 of a computer, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. Communication means 609 may allow electronic device 600 to communicate wirelessly or by wire with other devices to exchange data. Although FIG. 6 shows an electronic device 600 having various means, it should be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided.

For example, the processes described above with reference to the flowcharts may be implemented as computer software programs according to embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program comprising program code arranged to perform the method illustrated in the flowchart. In such an embodiment, the computer program may be downloaded and installed from the network via the communication device 609, or from the storage device 608, or from the ROM 602. When the computer program is executed by the processing apparatus 601, the above-mentioned functions defined in the methods of the embodiments of the present disclosure are executed.

It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples of computer readable storage media may include, but are not limited to, electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Programmable read only memory (EPROM or flash memory), fiber optics, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. In this disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present disclosure, however, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave with computer-readable program code embodied thereon. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device . Program code embodied on a computer readable medium may be transmitted using any suitable medium including, but not limited to, electrical wire, optical fiber cable, RF (radio frequency), etc., or any suitable combination of the foregoing.

In some embodiments, the client and server can use any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol) to communicate, and can communicate with digital data in any form or medium Communication (eg, a communication network) interconnects. Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet (eg, the Internet), and peer-to-peer networks (eg, ad hoc peer-to-peer networks), as well as any currently known or future development network of.

The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or may exist alone without being assembled into the electronic device.

The above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by the electronic device, the electronic device:

Determine the popularity level of the target media content;

Determine the target content distribution network CDN node type corresponding to the target media content based on the popularity level of the target media content; wherein the popularity level corresponds to the CDN node type one-to-one;

Computer program code for performing operations of the present disclosure may be written in one or more programming languages, including but not limited to object-oriented programming languages—such as Java, Smalltalk, C++, and This includes conventional procedural programming languages - such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider through Internet connection).

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more logical functions for implementing the specified functions executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented in dedicated hardware-based systems that perform the specified functions or operations , or can be implemented in a combination of dedicated hardware and computer instructions.

The units involved in the embodiments of the present disclosure may be implemented in a software manner, and may also be implemented in a hardware manner. Among them, the name of the unit does not constitute a limitation of the unit itself under certain circumstances.

The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), Systems on Chips (SOCs), Complex Programmable Logical Devices (CPLDs) and more.

In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in connection with the instruction execution system, apparatus or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media may include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), fiber optics, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

According to one or more embodiments of the present disclosure, a media content processing method, apparatus, device, and storage medium are provided, including:

Determine the popularity level of the target media content;

According to one or more embodiments of the present disclosure, a media content processing method, apparatus, device, and storage medium are provided, which determine the popularity level of target media content, including:

Based on the media content features of the target media content, a pre-trained tree model is used to predict the popularity level of the target media content, wherein the media content features include one or more of the following: media content visits, media content viewings , media content likes, media content comments, media content downloads, media content sharing.

According to one or more embodiments of the present disclosure, a media content processing method, apparatus, device, and storage medium are provided, wherein a pre-trained tree model is used to predict the popularity level of the target media content, including:

Utilize the pre-trained tree model to predict the relationship between the play growth amount of the target media content and the play growth amount threshold in a preset time period in the future;

The popularity level of the target media content is determined based on the relationship between the play increase amount of the target media content and the play increase amount threshold.

According to one or more embodiments of the present disclosure, a media content processing method, apparatus, device, and storage medium are provided, wherein the target media content is determined based on the relationship between a play increase amount of the target media content and a play increase amount threshold heat levels, including:

In response to determining that the play increase of the target media content is greater than or equal to the increase threshold, determining that the popularity level of the target media content is the first popularity;

In response to determining that the play increase of the target media content is less than the increase threshold, determining that the popularity level of the target media content is the second popularity.

According to one or more embodiments of the present disclosure, a media content processing method, apparatus, device, and storage medium are provided, further comprising:

According to one or more embodiments of the present disclosure, a media content processing method, apparatus, device, and storage medium are provided. The tree model is trained by using a binary cross-entropy loss function to obtain a pre-trained tree model, including:

Use multiple media content to build tree model training samples;

Determine the weight of each media content according to the media content characteristics of the plurality of media content, and determine the binary cross-entropy loss function according to the weight of each media content;

The tree model is trained based on the tree model training samples and the binary cross-entropy loss function to obtain a pre-trained tree model.

According to one or more embodiments of the present disclosure, a media content processing method, apparatus, device, and storage medium are provided, wherein the binary cross-entropy loss function is:

The foregoing descriptions are merely exemplary embodiments of the present disclosure and illustrative of the technical principles employed. Those skilled in the art should understand that the scope of the disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, and should also cover, without departing from the above-mentioned disclosed concept, the technical solutions formed by the above-mentioned technical features or Other technical solutions formed by any combination of its equivalent features. For example, a technical solution is formed by replacing the above features with the technical features disclosed in the present disclosure (but not limited to) with similar functions.

Additionally, although various operations are depicted in a particular order, this should not be construed as requiring that the operations be performed in the particular order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, although the above discussion contains several implementation-specific details, these should not be construed as limitations on the scope of the present disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.

Claims

A media content processing method, comprising:

Determine the popularity level of the target media content;

Determine the target content distribution network CDN node type corresponding to the target media content based on the popularity level of the target media content; wherein the popularity level corresponds to the CDN node type one-to-one;

Determine the target CDN node corresponding to the target media content based on the target CDN node type, and send the node information of the target CDN node to the client, so that the client can access the target CDN based on the node information The target media content is obtained from the node.
The method according to claim 1, wherein the determining the popularity level of the target media content comprises:

Based on the media content features of the target media content, a pre-trained tree model is used to predict the popularity level of the target media content, wherein the media content features include one or more of the following: media content visits, media content viewings , media content likes, media content comments, media content downloads, media content sharing.
The method according to claim 2, wherein the predicting the popularity level of the target media content by using a pre-trained tree model comprises:

Utilize the pre-trained tree model to predict the relationship between the play growth amount of the target media content and the play growth amount threshold in a preset time period in the future;

The popularity level of the target media content is determined based on the relationship between the play increase amount of the target media content and the play increase amount threshold.
The method according to claim 3, wherein the determining the popularity level of the target media content based on the relationship between the play increase amount of the target media content and the play increase amount threshold comprises:

In response to determining that the play increase of the target media content is greater than or equal to the increase threshold, determining that the popularity level of the target media content is the first popularity;

In response to determining that the play increase of the target media content is less than the increase threshold, determining that the popularity level of the target media content is the second popularity.
The method according to claim 2 or 3, further comprising:

The tree model is trained using a binary cross-entropy loss function to obtain the pre-trained tree model.
The method according to claim 5, wherein the tree model is trained by using a binary cross-entropy loss function to obtain the pre-trained tree model, comprising:

Use multiple media content to build tree model training samples;

Determine the weight of each media content according to the media content characteristics of the plurality of media content, and determine the binary cross-entropy loss function according to the weight of each media content;

The tree model is trained based on the tree model training samples and the binary cross-entropy loss function to obtain the pre-trained tree model.
The method of claim 6, wherein the binary cross-entropy loss function is:

Among them, N is the total amount of media content in the tree model training sample, yi represents the label of the ith media content in the tree model training sample, and pi represents the prediction of the ith media in the tree model training sample. The probability that the content is a positive example, α i represents the weight of the i-th media content in the tree model training sample,
vv represents the viewing volume of the ith media content sample in the tree model training sample in a preset time period in the future, VV represents the total viewing volume of the ith media content sample in the tree model training sample, and T represents the tree model The total duration corresponding to the total viewing amount of the ith media content sample in the training sample, t represents the viewing duration of the ith media content sample in the tree model training sample from the creation to the current time point.
A media content processing device, comprising:

The heat level determination module is set to determine the heat level of the target media content;

A node type determination module, configured to determine the target content distribution network CDN node type corresponding to the target media content based on the popularity level of the target media content, wherein the popularity level corresponds to the CDN node type one-to-one;

A target node determination module, configured to determine the target CDN node corresponding to the target media content based on the target CDN node type, and send the node information of the target CDN node to the client, so that the client is based on the target CDN node. The node information acquires the target media content from the target CDN node.
A media content processing device, comprising:

one or more processors;

memory, arranged to store one or more programs;

When the one or more programs are executed by the one or more processors, the one or more processors implement the media content processing method according to any one of claims 1-7.
A computer-readable storage medium storing a computer program, when the computer program is executed by a processor, implements the media content processing method according to any one of claims 1-7.