Multimedia information recommendation method and device, electronic equipment and storage medium

文档序号：1921701 发布日期：2021-12-03 浏览：8次中文

阅读说明：本技术 多媒体信息的推荐方法、装置、电子设备和存储介质 (Multimedia information recommendation method and device, electronic equipment and storage medium ) 是由杨晓宇卞俊杰叶璨于 2020-05-29 设计创作，主要内容包括：本公开关于一种多媒体信息的推荐方法、装置、电子设备和存储介质,所述多媒体信息的推荐方法包括：响应于接收到的多媒体信息推荐请求,获取发送所述多媒体信息推荐请求的账户的历史行为特征,其中,所述历史行为特征中至少包括所述账户与直播内容的交互信息；至少根据所述账户与直播内容的交互信息确定所述账户是否为直播活跃用户；在所述账户并非直播活跃用户的情况下,根据所述账户的基本特征和所述历史行为特征判断是否在返回多媒体资源的同时向所述账户返回直播聚合页,其中,所述历史行为特征还包括：所述账户与非直播内容的交互信息,所述直播聚合页用于汇集M个直播内容、并通过N个展示位展示所述直播聚合页,M、N为自然数,且M＞N。(The disclosure relates to a method and a device for recommending multimedia information, electronic equipment and a storage medium, wherein the method for recommending the multimedia information comprises the following steps: responding to a received multimedia information recommendation request, and acquiring historical behavior characteristics of an account sending the multimedia information recommendation request, wherein the historical behavior characteristics at least comprise interaction information of the account and live broadcast content; determining whether the account is a live active user or not according to at least interaction information of the account and live content; under the condition that the account is not a live broadcast active user, judging whether a live broadcast aggregated page is returned to the account while multimedia resources are returned according to the basic characteristics and the historical behavior characteristics of the account, wherein the historical behavior characteristics further comprise: the account is interactive with the non-live content, the live broadcast aggregated page is used for collecting M live broadcast contents and displaying the live broadcast aggregated page through N display positions, M, N is a natural number, and M is greater than N.)

1. A method for recommending multimedia information, the method comprising:

responding to a received multimedia information recommendation request, and acquiring historical behavior characteristics of an account sending the multimedia information recommendation request, wherein the historical behavior characteristics at least comprise interaction information of the account and live broadcast content;

determining whether the account is a live active user or not according to at least interaction information of the account and live content;

under the condition that the account is not a live broadcast active user, judging whether a live broadcast aggregated page is returned to the account while multimedia resources are returned according to the basic characteristics and the historical behavior characteristics of the account, wherein the historical behavior characteristics further comprise: the account is interactive with the non-live content, the live broadcast aggregated page is used for collecting M live broadcast contents and displaying the live broadcast aggregated page through N display positions, M, N is a natural number, and M is greater than N.

2. The method of claim 1, wherein returning the live aggregated page to the account comprises:

displaying a live broadcast aggregation entrance in a page of the application program logged in by the account;

and detecting the triggering operation of the live broadcast aggregation entrance, and jumping to a live broadcast aggregation page corresponding to the live broadcast aggregation entrance.

3. The method of claim 1, further comprising:

and returning the multimedia resources corresponding to the multimedia information recommendation request under the condition that the account is a live active user.

4. The method of claim 1, wherein the determining whether to return a live aggregated page to the account while returning multimedia resources according to the basic characteristics of the account and the historical behavior characteristics comprises:

vectorizing the basic features and the historical behavior features of the account to obtain account feature vectors corresponding to the basic features and the historical behavior features;

extracting the features of the account feature vector based on the trained deep network model;

and determining whether to return a live broadcast aggregation page to the account while returning multimedia resources according to the feature vectors extracted by the deep network model.

5. The method of claim 4, wherein the deep network model comprises a first network model and a second network model, the method further comprising:

determining the state s of the agent in the user according to an experience pool as a training sample_jExecute the recommended control action a_jPost-real-time Q-feedback r_jAccording to the instant feedback r_jDetermining the actual evaluation data y_j；

Based on the actual evaluation data y_jWith predictive assessment data Q(s) determined by a first one of the deep network models_j,a_j) The model parameters of the first network model are adjusted according to the difference, so that the deep network model containing the first network model after model parameter adjustment is determined as the trained deep network model.

6. The method of claim 5, wherein r is the instant feedback_jDetermining the actual evaluation data y_jThe method comprises the following steps:

detecting the user state s_jWith the next user state s_j+1The interval duration between is greater than the preset duration threshold value T_maxThen the instant feedback r is given_jAssigning to the actual evaluation data y_j；

Detecting the user state s_jWith the next user state s_j+1The interval duration between is not more than the preset duration threshold value T_maxAccording to the instant feedback r_jCorresponding to a next user state s derived from a second one of the deep network models_j+1Determining the actual evaluation data y_j。

7. The method of claim 1, wherein the determining whether the account is a live active user according to at least interaction information of the account and live content comprises:

and at least extracting the characteristics of the interactive information of the account and the live broadcast content based on a neural network regression model so as to determine whether the account is a live broadcast active account according to the extracted characteristics, wherein the neural network regression model is trained in advance based on a user information sample set, and the user information sample set at least comprises an interactive information sample of the live broadcast content and active degree marking information corresponding to the interactive information sample.

8. An apparatus for recommending multimedia information, said apparatus comprising:

the system comprises a characteristic acquisition module, a response module and a response module, wherein the response module is used for responding to a received multimedia information recommendation request and acquiring historical behavior characteristics of an account sending the multimedia information recommendation request, and the historical behavior characteristics at least comprise interactive information of the account and live broadcast content;

the user determination module is used for determining whether the account is a live active user or not at least according to the interaction information of the account and the live content;

an operation determining module, which determines whether to return a live aggregated page to the account while returning multimedia resources according to the basic characteristics of the account and the historical behavior characteristics when the account is not a live active user, wherein the historical behavior characteristics further include: the account is interactive with the non-live content, the live broadcast aggregated page is used for collecting M live broadcast contents and displaying the live broadcast aggregated page through N display positions, M, N is a natural number, and M is greater than N.

9. An electronic device, comprising:

a processor;

a memory for storing the processor-executable instructions;

wherein the processor is configured to execute instructions to implement the method of recommending multimedia information according to any of claims 1 to 7.

10. A storage medium, characterized in that instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform a method of recommending multimedia information according to any of claims 1 to 7.

Technical Field

The present disclosure relates to the field of internet, and in particular, to a method and an apparatus for recommending multimedia information, an electronic device, and a storage medium.

Background

With the rapid development of network technology and the application popularization of electronic equipment, user-oriented APP application develops rapidly, application functions become abundant, and people are full of application software with various functions and various forms in life, work and study. Meanwhile, in an increasingly rich application market, application function recommendation becomes a necessary means for improving market coverage of application functions.

However, in the related art, recommendation is performed for all users, a recommendation mode that does not distinguish a user group causes information interference to users who have the application function using habit, and even if recommendation is performed after the user group is distinguished, a recommendation rule set depending on manual experience in the related art is often limited by manual thinking dimension and subjective recognition, so that the identified user to be recommended is not accurate enough, and the recommendation efficiency is low.

Disclosure of Invention

The present disclosure provides a method and apparatus for recommending multimedia information, an electronic device, and a storage medium, to at least solve the technical problems in the related art. The technical scheme of the disclosure is as follows:

according to a first aspect of an embodiment of the present disclosure, a method for recommending multimedia information is provided, where the method includes:

determining whether the account is a live active user or not according to at least interaction information of the account and live content;

Optionally, the returning the live aggregated page to the account includes:

displaying a live broadcast aggregation entrance in a page of the application program logged in by the account;

and detecting the triggering operation of the live broadcast aggregation entrance, and jumping to a live broadcast aggregation page corresponding to the live broadcast aggregation entrance.

Optionally, the method further includes:

and returning the multimedia resources corresponding to the multimedia information recommendation request under the condition that the account is a live active user.

Optionally, the determining, according to the basic feature of the account and the historical behavior feature, whether to return a live aggregated page to the account while returning multimedia resources includes:

vectorizing the basic features and the historical behavior features of the account to obtain account feature vectors corresponding to the basic features and the historical behavior features;

extracting the features of the account feature vector based on the trained deep network model;

and determining whether to return a live broadcast aggregation page to the account while returning multimedia resources according to the feature vectors extracted by the deep network model.

Optionally, the deep network model includes a first network model and a second network model, and the method further includes:

determining the state s of the agent in the user according to an experience pool as a training sample_jExecute the recommended control action a_jLater immediate feedback r_jAccording to the instant feedback r_jDetermining the actual evaluation data y_j；

Optionally, said instant feedback r is based on_jDetermining the actual evaluation data y_jThe method comprises the following steps:

Optionally, the determining whether the account is a live active user according to at least the interaction information between the account and the live content includes:

Optionally, the determining whether the account is a live active account according to the extracted features includes:

determining the account as a feedback value of a live active user according to the extracted features; when the feedback value is detected to be lower than a preset feedback threshold value, determining that the account is a live broadcast inactive user; when the feedback value is detected to be not lower than a preset feedback threshold value, determining that the account is a live active user; alternatively, the first and second electrodes may be,

determining a first feedback value of the account as a live active user according to the extracted features, and determining a second feedback value of the account as the live active user; when the first feedback value is detected to be lower than the second feedback value, determining that the account is a live inactive user; and when the first feedback value is detected to be not lower than the second feedback value, determining that the account is a live active user.

According to a second aspect of the embodiments of the present disclosure, an apparatus for recommending multimedia information is provided, the apparatus comprising:

the user determination module is used for determining whether the account is a live active user or not at least according to the interaction information of the account and the live content;

Optionally, the operation determining module is specifically configured to:

displaying a live broadcast aggregation entrance in a page of the application program logged in by the account;

and detecting the triggering operation of the live broadcast aggregation entrance, and jumping to a live broadcast aggregation page corresponding to the live broadcast aggregation entrance.

Optionally, the method further includes:

and the resource returning module returns the multimedia resources corresponding to the multimedia information recommendation request under the condition that the account is a live active user.

Optionally, the operation determining module is further configured to:

vectorizing the basic features and the historical behavior features of the account to obtain account feature vectors corresponding to the basic features and the historical behavior features;

extracting the features of the account feature vector based on the trained deep network model;

and determining whether to return a live broadcast aggregation page to the account while returning multimedia resources according to the feature vectors extracted by the deep network model.

Optionally, the deep network model includes a first network model and a second network model, and the apparatus further includes:

an instant feedback determination module for determining the state s of the intelligent agent in the user according to the experience pool as the training sample_jExecute the recommended control action a_jLater immediate feedback r_jAccording to the instant feedback r_jDetermining the actual evaluation data y_j；

A model parameter adjustment module based on the actual evaluation data y_jWith predictive assessment data Q(s) determined by a first one of the deep network models_j,a_j) The model parameters of the first network model are adjusted according to the difference, so that the deep network model containing the first network model after model parameter adjustment is determined as the trained deep network model.

Optionally, the operation determining module is further configured to:

Detecting the user state s_jWith the next user state s_j+1The interval duration between is not more than the preset duration threshold value T_maxAccording to the instant feedback r_jCorresponding to a next use derived from a second one of the deep network modelsFamily state s_j+1Determining the actual evaluation data y_j。

Optionally, the user determination module is specifically configured to:

Optionally, the user determination module is further configured to:

According to a third aspect of the embodiments of the present disclosure, an electronic device is provided, including:

a processor;

a memory for storing the processor-executable instructions;

wherein the processor is configured to execute the instructions to implement the method for recommending multimedia information according to any of the above embodiments.

According to a fourth aspect of the embodiments of the present disclosure, a storage medium is provided, where instructions, when executed by a processor of an electronic device, enable the electronic device to perform the method for recommending multimedia information according to any of the embodiments described above.

According to a fifth aspect of the embodiments of the present disclosure, there is provided a computer program product configured to execute the method for recommending multimedia information according to any of the embodiments.

The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects:

according to the embodiment of the disclosure, under the condition that a multimedia information recommendation request is received, historical behavior characteristics of an account sending the multimedia information recommendation request can be acquired, whether the account is a live content active user or not is determined according to the acquired historical behavior characteristics, and further, under the condition that the account is not a live content active user, whether a live aggregation page is returned to the account while multimedia resources are returned is judged according to basic characteristics and historical behavior characteristics of the account, and through distinguishing whether the user is a live active user or not, information interference caused to users with live broadcast use habits due to live broadcast recommendation of the users can be reduced; in addition, whether live broadcast aggregation pages are returned to the account or not is determined according to the basic characteristics and the historical behavior characteristics of the account, the problems that recommendation efficiency is low and the like due to the fact that fixed recommendation rules set according to manual experience are recommended are solved, and recommendation efficiency of live broadcast recommendation performed on the account is improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.

Fig. 1 is a flowchart of a method for recommending multimedia information according to an exemplary embodiment of the present disclosure;

FIG. 2 is a flowchart of a deep network model training method for multimedia information recommendation provided according to an exemplary embodiment of the present disclosure;

FIG. 3 is a flowchart of another deep network model training method for multimedia information recommendation provided in accordance with an exemplary embodiment of the present disclosure;

FIG. 4 is a schematic diagram illustrating a live functionality recommendation in accordance with an illustrative embodiment of the present disclosure;

FIG. 5 is a schematic block diagram of an apparatus for recommending multimedia information according to one of the exemplary embodiments of the present disclosure;

fig. 6 is a schematic block diagram of a multimedia information recommendation apparatus according to a second exemplary embodiment of the present disclosure;

FIG. 7 is a schematic block diagram of a deep network model training apparatus for multimedia information recommendation shown in accordance with one of the illustrative embodiments of the present disclosure;

FIG. 8 is a schematic block diagram of a deep network model training apparatus for multimedia information recommendation according to a second exemplary embodiment of the present disclosure;

FIG. 9 is a schematic block diagram of a deep network model training apparatus for multimedia information recommendation according to another exemplary embodiment of the present disclosure;

FIG. 10 is a schematic block diagram of a deep network model training apparatus for multimedia information recommendation shown in the fourth embodiment of the present disclosure;

fig. 11 is a schematic block diagram illustrating an electronic device in accordance with an embodiment of the present disclosure.

Detailed Description

In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

The APP application can comprise application functions such as video recording, video editing, picture editing and short video sharing communities, each application function corresponds to an independent audience group or cross audiences exist among the application functions, recommendation information of the application functions is sent to non-audience groups of the application functions, so that the non-audience groups can contact the recommended application functions according to the recommendation information, even further know the recommended application functions deeply, and the number of the audience groups of the recommended application functions is increased.

Taking an exemplary live broadcast as an example, when the user a uses the application function of video recording, the access entry of other application functions, such as the aggregation entry of the live broadcast function, can be added in the product interface of video recording, so that the user can enter the product page of the live broadcast function through the aggregation entry of the live broadcast function to experience the live broadcast product function, thereby enabling the audience originally independent of the video recording application function to become the cross audience of the video recording function and the live broadcast function, and improving the user usage amount of the live broadcast function.

However, under the condition that a user has developed a good use habit related to a live broadcast function to be recommended, repeatedly recommending the live broadcast function to the user will certainly cause interference to the user, and interaction experience between the user and a product containing the live broadcast function is reduced, just as in the related art, if the live broadcast function is recommended to all known users, no matter whether the user has an application requirement on the live broadcast function to be recommended or not, recommendation information related to the live broadcast function will be received, information interference will be caused to the user having the use habit of the live broadcast function, and use experience of the user on the product containing the live broadcast function will be influenced; if the live broadcast function is recommended according to a set rule determined by artificial experience, the recommendation is limited by artificial thinking dimension or subjective cognition, so that the problems of inconsistent division of user groups, poor recommendation effect and the like are caused.

In view of the above, the present disclosure provides a method, an apparatus, an electronic device and a storage medium for recommending multimedia information, so as to solve at least the problems in the related art, and to illustrate the technical solutions of the present disclosure, the following describes the technical solutions of the present disclosure in detail through a plurality of embodiments.

Fig. 1 is a flowchart of a method for recommending multimedia information according to an exemplary embodiment of the present disclosure, and as shown in fig. 1, the method may include the following steps:

step 101, responding to a received multimedia information recommendation request, acquiring historical behavior characteristics of an account sending the multimedia information recommendation request, wherein the historical behavior characteristics at least comprise interaction information of the account and live broadcast content.

In an embodiment, in the case of receiving a multimedia information recommendation request, historical behavior characteristics of an account sending the multimedia information recommendation request may be obtained, where the historical behavior characteristics at least include interaction information of the account and live content.

Specifically, at least the interactive information between the account and the live content can be written in the multimedia information recommendation request, so that the interactive information between the account and the live content contained in the request information can be at least determined by analyzing the received request information; user identification information such as a user account, a user identity certificate and the like can also be determined according to the received request information, and then interactive information matched with the user identification information is determined in the association relationship between the maintained user identification information and the interactive information of the account and the live broadcast content.

Further, besides the interactive information of the account and the live broadcast content, the basic information of the user can be included, such as the number of people concerned by the user, the number of fans, the age, the sex, the region, the number of active days and other information; user behavior information such as the number of exposures, the number of clicks, the length of use time, and the like with respect to a preset live broadcast; context information such as exposure rate of a preset live broadcast within a past period of time, click rate, duration of entering a live broadcast, etc. In the present disclosure, several types of user basic information, user behavior information, and context information are listed by way of example, the present disclosure does not limit the specific form of information specifically included in various information categories, and other information categories are added according to the actual application requirements except for the user basic information, the user behavior information, and the context information, and the present disclosure does not limit this.

And step 102, determining whether the account is a live active user or not at least according to the interaction information of the account and the live content.

In an embodiment, vectorization processing may be performed on the basic features and the historical behavior features of the account to obtain account feature vectors corresponding to the basic features and the historical behavior features of the account, feature extraction may be performed on the account feature vectors based on the trained deep network model, and then it is determined whether to return a live aggregation page to the account while returning multimedia resources according to the feature vectors extracted by the deep network model.

Specifically, the Deep Network model may be a mathematical model used for simulating a sequential decision class of randomness policy and return that can be achieved by an agent in an environment with a markov property, such as a Double Deep Q Network model (DDQN), and the account feature vector is subjected to feature extraction by the trained Deep Network model, so as to determine whether to return a live aggregation page to the account while returning multimedia resources according to the feature vector extracted by the Deep Network model.

In practical application, the deep network model may include a first network model and a second network model, and the deep network model for feature extraction is trainedThe refining process can comprise the following processes: determining the state s of the agent in the user according to an experience pool as a training sample_jExecute the recommended control action a_jLater immediate feedback r_jAccording to the instant feedback r_jDetermining the actual evaluation data y_j(ii) a Further, based on the actual evaluation data y_jWith predictive assessment data Q(s) determined by a first one of the deep network models_j,a_j) The model parameters of the first network model are adjusted according to the difference, so that the deep network model containing the first network model after model parameter adjustment is determined as the trained deep network model.

Further, based on the instant feedback r_jDetermining the actual evaluation data y_jThe process of (a) may include: detecting the user state s_jWith the next user state s_j+1The interval duration between is greater than the preset duration threshold value T_maxThen the instant feedback r is given_jAssigning to the actual evaluation data y_j(ii) a Detecting the user state s_jWith the next user state s_j+1The interval duration between is not more than the preset duration threshold value T_maxAccording to the instant feedback r_jCorresponding to a next user state s derived from a second one of the deep network models_j+1Determining the actual evaluation data y_j。

Before vectorization processing is performed on the user state information of the features to be extracted, word segmentation processing can be performed on the basic features and the historical behavior features of the account, and then vectorization processing is performed according to word information after word segmentation processing. Specifically, word vectors corresponding to the basic information and the historical behavior features of the account may be determined by a co-occurrence matrix, singular value decomposition, or the like, or word vectors corresponding to the basic information and the historical behavior features of the account may be determined by a language model such as CBOW.

In this embodiment, before feature extraction is performed on user state information based on the reinforcement learning model, the basic information and the historical behavior features of the account may be preprocessed, and vectorization processing may be performed on the basic information and the historical behavior features of the account whose features are to be extracted, so as to improve the feature extraction efficiency of the reinforcement learning model.

In another embodiment, before feature extraction is performed on at least the interaction information of the account and the live content based on the deep network model, whether the account is a live active account or not can be determined according to the extracted features, so that feature extraction is performed on the basic features and the historical behavior features of the account of the live inactive user, and feature extraction is not performed on the live active user, so that the overall processing efficiency is improved.

Specifically, feature extraction is performed on at least the interactive information of the account and the live broadcast content based on a neural network regression model, so as to determine whether the account is a live broadcast active account or not according to the extracted features, wherein the neural network regression model is trained in advance based on a user information sample set, and the user information sample set at least comprises an interactive information sample of the live broadcast content and activity degree marking information corresponding to the interactive information sample.

Further, in the process of determining whether the account is a live active account according to the extracted features, a feedback value of the account as a live active user can be determined according to the extracted features, when the feedback value is detected to be lower than a preset feedback threshold value, the account is determined to be a live inactive user, and when the feedback value is detected to be not lower than the preset feedback threshold value, the account is determined to be a live active user; or determining the account as a first feedback value of a live active user according to the extracted features, and determining the account as a second feedback value of the live active user; when the first feedback value is detected to be lower than the second feedback value, determining that the account is a live inactive user; and when the first feedback value is detected to be not lower than the second feedback value, determining that the account is a live active user.

In this embodiment, at least feature extraction may be performed on the interaction information of the account and the live content to determine the usage probability of the account for live broadcast, and then it is determined whether the account is a live active user according to the usage probability of the account for live broadcast. In practical application, the interactive information with the live content may include historical interactive behavior with the live content and context information interacting with the live content, and may also perform feature extraction on other content except the interactive information between the account and the live content, and the specific content may be set according to practical application conditions.

103, under the condition that the account is not a live broadcast active user, judging whether a live broadcast aggregation page is returned to the account while multimedia resources are returned according to the basic characteristics of the account and the historical behavior characteristics, wherein the historical behavior characteristics further comprise: the account is interactive with the non-live content, the live broadcast aggregated page is used for collecting M live broadcast contents and displaying the live broadcast aggregated page through N display positions, M, N is a natural number, and M is greater than N.

In this embodiment, the live broadcast aggregation entry may be displayed in a page of an application program in which an account logs in, and when a trigger operation on the live broadcast aggregation entry is detected, a jump is made to a live broadcast aggregation page corresponding to the live broadcast aggregation entry. In this embodiment, the live broadcast aggregation entry is displayed in the page of the application program logged in by the account instead of directly returning to the live broadcast aggregation page, so that other contents except for the live broadcast aggregation entry can be displayed to the user in addition to the live broadcast aggregation entry returned to the user, and further, whether the live broadcast aggregation entry is returned or not can be determined according to whether a trigger operation on the live broadcast aggregation entry is received, so that the interaction experience of the user is enhanced.

In this embodiment, the multimedia resource corresponding to the multimedia information recommendation request may be returned when the account is a live active user. When the account is detected to be a live broadcast active user, the multimedia resource corresponding to the multimedia information recommendation request is directly returned, and a live broadcast aggregation page is not returned to the account, so that the interference to the original live broadcast active user due to the return of the live broadcast aggregation page is avoided.

According to the embodiment, under the condition that the multimedia information recommendation request is received, historical behavior characteristics of an account sending the multimedia information recommendation request can be acquired, whether the account is a live content active user or not is determined according to the acquired historical behavior characteristics, whether a live aggregation page is returned to the account while multimedia resources are returned or not is judged according to basic characteristics and historical behavior characteristics of the account under the condition that the account is not a live active user, and information interference caused to users with live use habits due to live recommendation of the users can be reduced by distinguishing whether the users are live active users or not; in addition, whether live broadcast aggregation pages are returned to the account or not is determined according to the basic characteristics and the historical behavior characteristics of the account, the problems that recommendation efficiency is low and the like due to the fact that fixed recommendation rules set according to manual experience are recommended are solved, and recommendation efficiency of live broadcast recommendation performed on the account is improved.

Fig. 2 is a flowchart of a deep network model training method for multimedia information recommendation according to an exemplary embodiment of the present disclosure, where as shown in fig. 2, the deep network model may include a dual deep Q network, and the training method may include the following steps:

step 201, initializing a current network Q and a target network Q in a double-depth Q network^*The network model of (1).

In one embodiment, the current network Q and the target network Q in the initialized dual-depth Q network^*May have the same network structure, such as the same number of network layers, consistent parameters of the network model, etc. Specifically, a current network Q and a target network Q in a pair dual-depth Q network^*In the process of initialization, the current network Q and the target network Q can be determined in a random assignment mode^*Such as gaussian random assignments, etc., which the present disclosure does not limit.

Step 202, determining a user state s containing the agent according to the experience pool as a training sample_jPerforming an actual recommended control action a_jImmediate reward r_jThe sample data of (1).

Step 203, according to the instant prize r_jDetermining an actual value estimate y_j。

In an embodiment, the user state s may be detected based on_jWith the lower partA user state s_j+1Interval duration between and preset duration threshold T_maxRelation between them, determining the actual value evaluation value y based on different ways_jAnd the user state at least comprises basic characteristics and historical behavior characteristics of the account, and the historical behavior characteristics comprise interaction information of the account and the non-live content.

In particular, when the user state s is detected_jWith the next user state s_j+1The interval duration between is greater than the preset duration threshold value T_maxIn case of (2), the award r is given_jIs assigned to the actual value evaluation value y_j(ii) a Upon detection of said user state s_jWith the next user state s_j+1The interval duration between is not more than the preset duration threshold value T_maxAccording to the award r_jAnd by the target network Q^*The obtained state s corresponding to the next user_j+1Determines the actual value estimate y_j。

In the present embodiment, the determination manner of the actual value evaluation value is determined according to the interval duration between the user state and the next user state, so that in the case where the interval duration between the user state and the next user state is short, the actual value evaluation value is determined more from the value evaluation values obtained by the target network corresponding to the next user state, and in the case where the interval duration between the user state and the next user state is long, the actual value evaluation value is determined from the received immediate reward, and the accuracy and the practicality of the process of determining the actual value evaluation value are realized.

Further, based on the award r_jAnd by the target network Q^*The obtained state s corresponding to the next user_j+1Determines an actual value evaluation value y_jCan be based on the user status s_jWith the next user state s_j+1Interval duration between T(s)_j,s_j+1) Determining a discount coefficient alpha_jThe discount coefficient α_j＝γexp(-T(s_j,s_j+1)/T_max) (ii) a Further according to the reward r_jDiscount coefficient alpha_jAnd by the target network Q^*The obtained state s corresponding to the next user_j+1Determines the actual value estimate y_jWherein the actual value evaluation value

In the present embodiment, a further improvement is made to the manner of determining the actual value evaluation value by using the fixed discount coefficient in the related art, that is, changing the original fixed discount coefficient into the dynamic discount coefficient α related to the interval duration between the user state and the next user state_jIn the case that the interval duration between the user state and the next user state is longer, the discount coefficient is dynamically changed to a smaller value, so that the influence of the value evaluation value of the next user state is less considered, and the influence of the instant prize on the actual value evaluation value is more considered; and under the condition that the interval duration between the user state and the next user state is short, the discount coefficient is dynamically changed into a larger numerical value, so that the influence of the value evaluation value of the next user state is considered more, the setting of the dynamic discount coefficient enables the determination process of the actual value evaluation value to be matched with the actual thinking habits of the user, the effectiveness and the practicability of the determination of the actual value evaluation value are realized, and the optimization efficiency of the model parameters is improved.

Step 204, based on the actual value evaluation value y_jWith a predicted value estimate Q(s) determined by the current network Q_j,a_j) Adjusts the model parameters of the current network Q.

In an embodiment, the following iteration process may be repeated until the number of iterations reaches a preset number threshold, and then the model parameter of the current network Q is assigned to the target network Q^*: and acquiring the next user state after the intelligent agent executes the actual recommended control action, and adjusting the model parameters of the current network Q according to the difference between the actual value evaluation value and the predicted value evaluation value determined by the sample data corresponding to the next user state.

In this embodiment, after a preset number of iterations, model parameters of the current network Q are assigned to the target network Q^*Make the target network Q^*The model parameter updating process of (1) keeps certain noise, and avoids the target network Q participating in error analysis due to overfitting^*And distortion improves the effectiveness of model parameter training of the current network Q.

And step 205, performing feature extraction on at least basic features and historical behavior features of the account according to the trained dual-depth Q network so as to implement live recommendation control according to the optimal action determined by the extracted features.

In one embodiment, the value y may be estimated based on the actual value_jWith a predicted value estimate Q(s) determined by the current network Q_j,a_j) Mean square error loss value between to determine the actual value evaluation value y_jWith a predicted value estimate Q(s) determined by the current network Q_j,a_j) The difference between them. Specifically, the actual value evaluation value y_jWith a predicted value estimate Q(s) determined by the current network Q_j,a_j) The mean square error loss value between can be determined by a loss functionAnd is determined.

Actual value estimate y determined by back propagation_jWith a predicted value estimate Q(s) determined by the current network Q_j,a_j) The difference between the two parameters enables the double-depth Q network to optimize the model parameters of the double-depth Q network based on the gradient descent method until the determined actual value evaluation value y_jWith a predicted value estimate Q(s) determined by the current network Q_j,a_j) And when the difference between the two parameters is lower than a difference threshold value, determining the double-depth Q network after model parameter adjustment as the double-depth Q network after training.

As can be seen from the above embodiments, the dual-depth Q network for determining the optimal action according to the characteristics of the user state information may be previously defined by the intelligent agent in the user state s_jExecuting actual recommended controlsAs a_jImmediate reward r_jThe actual value evaluation value y is obtained by training the sample data in the experience pool in the training process_jThe model parameter of the current network Q is adjusted according to the difference between the determined instant reward and the predicted value assessment value determined by the current network Q, and the model parameter of the current network Q is corrected in time to be matched with the determined actual value assessment value, so that the updating efficiency of model training is improved; in addition, the double-depth Q network trained in advance is used for analyzing according to the user state, and then determining the optimal action related to live broadcast recommendation control, so that the determination efficiency of the optimal action is improved, the problem that the accuracy of the identified user group to be recommended is low due to the limitation of artificial thinking dimension or subjective understanding is avoided, and the live broadcast recommendation effectiveness is improved.

In the present disclosure, whether an account is a live active user may be determined according to a neural network model that is trained in advance, a training process of the neural network model may be as shown in fig. 3, where fig. 3 is a flowchart of another deep network model training method for multimedia information recommendation provided according to an exemplary embodiment of the present disclosure, and the following steps may be specifically involved in the training process of the neural network model:

step 301, determining an active information sample set as a training sample, where the active information sample set includes user state information and live broadcast activity degree labeling information corresponding to the user state information.

In an embodiment, the user status information may include at least one of the following information of the determined active information sample set as the training sample: the method comprises the steps that basic characteristics and historical behavior characteristics of an account are obtained, wherein the historical behavior characteristics at least comprise interaction information of the account and live content, interaction information of the account and non-live content and the like, an active information sample set can be composed of user state information and live activity degree marking information corresponding to the user state information, for example, when the user state information is user basic information a, the user behavior information b and the context information c, a user account corresponding to the user state information which is live broadcast is an active user, and exemplarily, live activity degree marking information corresponding to the user state information is an active state; and when the user state information is the user basic information d, the user behavior information e, and the context information f, the user account corresponding to the live broadcast user state information is an inactive user, and exemplarily, the live broadcast activity degree marking information corresponding to the user state information may be an inactive state.

Step 302, performing feature extraction on the user state information by the neural network model to determine activity prediction information according to the extracted features.

In an embodiment, vectorization processing may be performed on the user state information to obtain a user state vector corresponding to the user state information, and then feature extraction is performed on the user state vector by using a neural network model to determine activity degree prediction information according to the extracted features, where the activity degree prediction information represents a probability value p that a user account corresponding to the live broadcast user state information is in an active state.

Step 303, determining a difference between the activity level labeling information and the activity level prediction information, so as to adjust a model parameter of a neural network model according to the difference of the back propagation.

In an embodiment, the difference between the activity level labeling information and the activity level prediction information may be determined by a loss function corresponding to the neural network model. Specifically, the loss function L may be L ═ x log (p) - (1-x) log (1-p), where x represents activity level labeling information used for indicating whether a user account corresponding to the user state information input to the neural network model is actually an active user; and p represents the probability p of the account in the active state, which is predicted according to the extracted features after the neural network model extracts the features of the user state information, namely the activity degree prediction information p.

Further, after determining the loss value between the activity degree labeling information and the activity degree prediction information based on the loss function, the determined loss value can be propagated reversely, and then the neural network model parameters are optimized according to a gradient descent method until the loss value between the activity degree labeling information and the activity degree prediction information is lower than a preset loss threshold value based on the loss function, and the neural network model after model parameter adjustment is determined to be a trained regression model.

Fig. 4 is a schematic diagram of a live broadcast function recommendation shown in the present disclosure according to an exemplary embodiment, as shown in fig. 4, after a multimedia information recommendation request is detected, feature extraction may be performed on historical behavior features of an account sending the multimedia information recommendation request through a neural network model, where the historical behavior features at least include interaction information of the account and live broadcast content, so as to determine whether the account is a live active user according to the extracted features, specifically, basic features and historical behavior features of the account may be provided to the neural network model, so as to determine whether the account is a live active user according to the extracted features by the neural network model.

Specifically, in the process of determining whether the account is a live broadcast active account according to the features extracted by the neural network model, in an embodiment, the account may be determined as a feedback value of a live broadcast active user according to the extracted features; when the feedback value is detected to be lower than a preset feedback threshold value, determining that the account is a live broadcast inactive user; and when the feedback value is detected to be not lower than a preset feedback threshold value, determining that the account is a live active user.

In another embodiment, a first feedback value that the account is a live active user and a second feedback value that the account is a live active user may be determined according to the extracted features; when the first feedback value is lower than the second feedback value, determining that the account is a live broadcast inactive user; and when the first feedback value is detected to be not lower than the second feedback value, determining that the account is a live active user.

If the account is determined to be a live active user by the neural network model, returning the multimedia resource corresponding to the multimedia information recommendation request; if the account is determined to be not a live active user by the neural network model, at least the basic features and the historical behavior features of the account are provided to the deep network model, so as to determine whether to return a live aggregated page to the account while returning multimedia resources according to at least the basic features and the historical behavior features of the account, wherein the deep network model may include the pre-trained dual-deep Q network mentioned in the above embodiment.

In practical application, the process of returning the live broadcast aggregation page to the account can be to display a live broadcast aggregation entry in a page of an application program logged in by the account; and then detecting the triggering operation of the live broadcast aggregation entrance, and jumping to a live broadcast aggregation page corresponding to the live broadcast aggregation entrance.

While, for purposes of simplicity of explanation, the foregoing method embodiments have been described as a series of acts or combination of acts, it will be appreciated by those skilled in the art that the present disclosure is not limited by the order of acts, as some steps may, in accordance with the present disclosure, occur in other orders and concurrently.

Further, those skilled in the art will appreciate that the embodiments described in the specification are all alternative embodiments.

The present disclosure also proposes an embodiment of an image editing apparatus, corresponding to the aforementioned embodiment of the image editing method.

Fig. 5 is a schematic block diagram of a recommendation apparatus for multimedia information according to one of the exemplary embodiments of the present disclosure. The recommendation apparatus for multimedia information shown in this embodiment may be applied to a video playing application, where the application is applied to a terminal, and the terminal includes, but is not limited to, a mobile phone, a tablet computer, a wearable device, a personal computer, and other electronic devices. The video playing application may be an application installed in the terminal, or may be a web application integrated in the browser, and the user may play a video through the video playing application, where the played video may be a long video, such as a movie and a tv series, or a short video, such as a video clip and a scene short series.

Referring to fig. 5, the apparatus may include a feature acquisition module 501, a user determination module 502, an operation determination module 503; wherein the content of the first and second substances,

the feature obtaining module 501, configured to obtain, in response to a received multimedia information recommendation request, historical behavior features of an account that sends the multimedia information recommendation request, where the historical behavior features at least include interaction information between the account and live content;

a user determining module 502, configured to determine whether the account is a live active user at least according to interaction information between the account and live content;

an operation determining module 503, when the account is not a live active user, determining whether to return a live aggregated page to the account while returning multimedia resources according to the basic characteristics of the account and the historical behavior characteristics, where the historical behavior characteristics further include: the account is interactive with the non-live content, the live broadcast aggregated page is used for collecting M live broadcast contents and displaying the live broadcast aggregated page through N display positions, M, N is a natural number, and M is greater than N.

Optionally, the operation determining module 503 is specifically configured to:

displaying a live broadcast aggregation entrance in a page of the application program logged in by the account;

and detecting the triggering operation of the live broadcast aggregation entrance, and jumping to a live broadcast aggregation page corresponding to the live broadcast aggregation entrance.

Optionally, the operation determining module 503 is further configured to:

vectorizing the basic features and the historical behavior features of the account to obtain account feature vectors corresponding to the basic features and the historical behavior features;

extracting the features of the account feature vector based on the trained deep network model;

and determining whether to return a live broadcast aggregation page to the account while returning multimedia resources according to the feature vectors extracted by the deep network model.

Optionally, the deep network model includes a first network model and a second network model, and the operation determining module 503 is further configured to:

an immediate feedback determination module based onDetermining agent state s as a user state of a training sample_jExecute the recommended control action a_jLater immediate feedback r_jAccording to the instant feedback r_jDetermining the actual evaluation data y_j；

Optionally, the operation determining module 503 is further configured to:

Optionally, the user determination module 502 is specifically configured to:

Optionally, the user determination module 502 is further configured to:

As shown in fig. 6, fig. 6 is a schematic block diagram of a multimedia information recommendation apparatus according to a second exemplary embodiment of the present disclosure, and the embodiment may further include a resource returning module 504 on the basis of the foregoing embodiment shown in fig. 5:

and the resource returning module 504 is used for returning the multimedia resource corresponding to the multimedia information recommendation request under the condition that the account is a live active user.

Fig. 7 is a schematic block diagram of a deep network model training apparatus for multimedia information recommendation according to one of the exemplary embodiments of the present disclosure. The deep network model training device for multimedia information recommendation shown in this embodiment may be applied to video playing applications, where the applications are applied to terminals, and the terminals include, but are not limited to, mobile phones, tablet computers, wearable devices, personal computers, and other electronic devices. The video playing application may be an application installed in the terminal, or may be a web application integrated in the browser, and the user may play a video through the video playing application, where the played video may be a long video, such as a movie and a tv series, or a short video, such as a video clip and a scene short series.

Referring to fig. 7, the apparatus may include a model initialization module 701, a sample data determination module 702, an evaluation value determination module 703, a first parameter adjustment module 704, a first feature extraction module 705; wherein the content of the first and second substances,

model initialization module 701 for initializing current network Q and target network Q in dual-depth Q network^*The network model of (2);

a sample data determining module 702 for determining the state s of the intelligent agent in the user according to the experience pool as the training sample_jPerforming an actual recommended control action a_jImmediate reward r_jThe sample data of (1);

a valuation determination module 703 for determining the real-time award r_jDetermining an actual value estimate y_j；

A first parameter adjusting module 704 for adjusting the actual value y based on the actual value_jWith a predicted value estimate Q(s) determined by the current network Q_j,a_j) Adjusts the model parameters of the current network Q;

the first feature extraction module 705 performs feature extraction on at least basic features and historical behavior features of the account according to the trained dual-depth Q network so as to implement recommendation control on live broadcast according to the optimal action determined by the extracted features.

Optionally, the method further includes:

and a model parameter determination module 712, determining the model parameter when the loss value is lower than a preset threshold value as the model parameter of the trained neural network model.

As shown in fig. 8, fig. 8 is a schematic block diagram of a deep network model training apparatus for multimedia information recommendation according to a second exemplary embodiment of the present disclosure, where the embodiment may further include, on the basis of the foregoing embodiment shown in fig. 7: a state obtaining module 706, a second parameter adjusting module 707, and a parameter assigning module 708; wherein the content of the first and second substances,

the state obtaining module 706 obtains the next user state after the agent executes the actual recommended control action;

a second parameter adjusting module 707 for adjusting a model parameter of the current network Q according to a difference between an actual value evaluation value and a predicted value evaluation value determined from sample data corresponding to the next user state;

parameter assignment module 708, assigning the model parameters of the current network Q to the target network Q under the condition that the times of repeatedly executing the two steps reach a preset time threshold value^*。

As shown in fig. 9, fig. 9 is a schematic block diagram of a deep network model training apparatus for multimedia information recommendation according to a third exemplary embodiment of the present disclosure, where the embodiment is based on the foregoing embodiment shown in fig. 7, and may further include: a sample set determination module 709, a second feature extraction module 710, and a third parameter adjustment module 711; wherein the content of the first and second substances,

a sample set determining module 709, configured to determine an active information sample set as a training sample, where the active information sample set includes user status information and live activity tagging information corresponding to the user status information;

a second feature extraction module 710, for performing feature extraction on the user status information by the neural network model to determine activity prediction information according to the extracted features;

a third parameter adjusting module 711, configured to determine a difference between the activity labeling information and the activity prediction information, so as to adjust a model parameter of the neural network model according to the difference propagated in a backward direction.

As shown in fig. 10, fig. 10 is a schematic block diagram of a deep network model training apparatus for multimedia information recommendation according to a fourth exemplary embodiment of the present disclosure, where on the basis of the foregoing embodiment shown in fig. 7, the evaluation value determining module 703 may include: an assignment sub-module 7031, an evaluation value determination sub-module 7032; wherein the content of the first and second substances,

an assignment submodule 7031 for assigning said user state s upon detection of said user state s_jWith the next user state s_j+1The interval duration between is greater than the preset duration threshold value T_maxIn case of (2), the instant prize r is given_jIs assigned to the actual value evaluation value y_j；

Evaluation value determining sub-module 7032 for determining the evaluation value when the user state s is detected_jWith the next user state s_j+1The interval duration between is not more than the preset durationDuration threshold T_maxAccording to the award r_jAnd by the target network Q^*The obtained state s corresponding to the next user_j+1Determines the actual value estimate y_j。

For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. One of ordinary skill in the art can understand and implement it without inventive effort.

An embodiment of the present disclosure also provides an electronic device, including:

a processor;

a memory for storing the processor-executable instructions;

wherein the processor is configured to execute the instructions to implement the method for recommending multimedia information according to any of the above embodiments.

Embodiments of the present disclosure also provide a storage medium, where instructions executed by a processor of an electronic device enable the electronic device to perform the method for recommending multimedia information according to any of the above embodiments.

Embodiments of the present disclosure further provide a computer program product configured to execute the method for recommending multimedia information according to any of the above embodiments.

Fig. 11 is a schematic block diagram illustrating an electronic device in accordance with an embodiment of the present disclosure. For example, the electronic device 1100 can be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.

Referring to fig. 11, electronic device 1100 may include one or more of the following components: processing component 1102, memory 1104, power component 1106, multimedia component 1108, audio component 1110, input/output (I/O) interface 1113, sensor component 1114, and communications component 1116.

The processing component 1102 generally controls the overall operation of the electronic device 1100, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 1102 may include one or more processors 1120 to execute instructions to perform all or a portion of the steps of the method for recommending multimedia information described above. Further, the processing component 1102 may include one or more modules that facilitate interaction between the processing component 1102 and other components. For example, the processing component 1102 may include a multimedia module to facilitate interaction between the multimedia component 1108 and the processing component 1102.

The memory 1104 is configured to store various types of data to support operations at the electronic device 1100. Examples of such data include instructions for any application or method operating on the electronic device 1100, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 1104 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

The power supply component 1106 provides power to the various components of the electronic device 1100. The power components 1106 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device 1100.

The multimedia component 1108 includes a screen that provides an output interface between the electronic device 1100 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 1108 includes a front facing camera and/or a rear facing camera. The front camera and/or the rear camera may receive external multimedia data when the electronic device 1100 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

The audio component 1110 is configured to output and/or input audio signals. For example, the audio component 1110 includes a Microphone (MIC) configured to receive external audio signals when the electronic device 1100 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 1104 or transmitted via the communication component 1116. In some embodiments, the audio assembly 1110 further includes a speaker for outputting audio signals.

The I/O interface 1113 provides an interface between the processing component 1102 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.

The sensor assembly 1114 includes one or more sensors for providing various aspects of state assessment for the electronic device 1100. For example, the sensor assembly 1114 may detect an open/closed state of the electronic device 1100, the relative positioning of components, such as a display and keypad of the electronic device 1100, the sensor assembly 1114 may also detect a change in the position of the electronic device 1100 or a component of the electronic device 1100, the presence or absence of user contact with the electronic device 1100, orientation or acceleration/deceleration of the electronic device 1100, and a change in the temperature of the electronic device 1100. The sensor assembly 1114 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 1114 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 1114 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 1116 is configured to facilitate wired or wireless communication between the electronic device 1100 and other devices. The electronic device 1100 may access a wireless network based on a communication standard, such as WiFi, a carrier network (such as 2G, 3G, 4G, or 5G), or a combination thereof. In an exemplary embodiment, the communication component 1116 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 1116 also includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an embodiment of the present disclosure, the electronic device 1100 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-mentioned recommendation method of multimedia information.

In an embodiment of the present disclosure, a non-transitory computer-readable storage medium including instructions, such as the memory 1104 including instructions, which are executable by the processor 1120 of the electronic device 1100 to perform the method for recommending multimedia information is also provided. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The method and apparatus provided by the embodiments of the present disclosure are described in detail above, and the principles and embodiments of the present disclosure are explained herein by applying specific examples, and the above description of the embodiments is only used to help understanding the method and core ideas of the present disclosure; meanwhile, for a person skilled in the art, based on the idea of the present disclosure, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present disclosure should not be construed as a limitation to the present disclosure.

25页详细技术资料下载

Multimedia information recommendation method and device, electronic equipment and storage medium

相关技术

网友询问留言