CN116128324A - Multimedia resource evaluation and pushing method and server - Google Patents

Multimedia resource evaluation and pushing method and server Download PDF

Info

Publication number
CN116128324A
CN116128324A CN202111341553.XA CN202111341553A CN116128324A CN 116128324 A CN116128324 A CN 116128324A CN 202111341553 A CN202111341553 A CN 202111341553A CN 116128324 A CN116128324 A CN 116128324A
Authority
CN
China
Prior art keywords
multimedia resource
information
characteristic
feedback information
value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111341553.XA
Other languages
Chinese (zh)
Inventor
田靖玉
赵耀红
遇倩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Mobile Communications Group Co Ltd
China Mobile Communications Ltd Research Institute
Original Assignee
China Mobile Communications Group Co Ltd
China Mobile Communications Ltd Research Institute
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Mobile Communications Group Co Ltd, China Mobile Communications Ltd Research Institute filed Critical China Mobile Communications Group Co Ltd
Priority to CN202111341553.XA priority Critical patent/CN116128324A/en
Publication of CN116128324A publication Critical patent/CN116128324A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/45—Clustering; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking

Abstract

The disclosure provides a method for evaluating and pushing multimedia resources and a server, wherein the method comprises the following steps: acquiring characteristic values of information of the multimedia resources, and determining an evaluation result of the multimedia resources according to the characteristic values of the information and the characteristic values of user feedback information corresponding to the multimedia resources. By the method, the multimedia resources can be accurately evaluated.

Description

Multimedia resource evaluation and pushing method and server
Technical Field
The invention relates to the technical field of data processing, in particular to a method for evaluating and pushing multimedia resources and a server side.
Background
The existing evaluation method of the multimedia resources mainly comprises manual evaluation, wherein the manual evaluation can be divided into expert panel evaluation and public panel evaluation under various evaluation occasions, for example, for a certain multimedia resource, a viewer can score according to own preference after watching the multimedia resource, and after a period of time, the average score of all users for the multimedia resource is obtained, and the average score is used as the evaluation score of the multimedia resource.
Therefore, the evaluation of the multimedia resources is mainly based on subjective evaluation factors of the users, and the subjective evaluation factors of the users are influenced by moods, environments and the like of the users during evaluation, so that the evaluation scores lack objectivity and accuracy.
Disclosure of Invention
At least one embodiment of the present disclosure provides a method for evaluating and pushing multimedia resources and a server,
according to one aspect of the present disclosure, at least one embodiment provides a method for evaluating a multimedia resource, the method comprising:
acquiring a characteristic value of information of the multimedia resource;
and determining an evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource.
In this embodiment, the multimedia resource may be at least one multimedia resource in a preset time window.
The feature value of the information of the multimedia resource may be obtained by obtaining the feature value of the information of at least one multimedia resource in a predetermined time window.
In this embodiment, the information of the multimedia resource is used as one of the conditions for forming the evaluation result of the multimedia resource, so that the subjective evaluation of the user and the objective evaluation of the multimedia resource can be used to comprehensively evaluate the multimedia resource, and the evaluation result of the multimedia resource is more objective and accurate.
In addition, the embodiment of the disclosure also provides a method for obtaining the characteristic value of the information of the multimedia resource, which comprises the following steps:
and acquiring personnel characteristic information and/or video characteristic information of the multimedia resource, and acquiring characteristic values of information according to quantization results of the personnel characteristic information and/or the video characteristic information.
Wherein, the obtaining the characteristic value of the information according to the quantization result of the personnel characteristic information and/or the video characteristic information comprises the following steps:
and obtaining the characteristic value of the information according to the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the personnel characteristic information and/or the video characteristic information.
In one embodiment, the normalized result of the encoding result of the discrete feature and the derived result of the continuous feature in the personnel feature information and/or the video feature information is used as the feature value of the information.
In addition, the embodiment of the disclosure further provides a method for obtaining the characteristic value of the user feedback information corresponding to the multimedia resource, which comprises the following steps:
and acquiring objective feedback information and/or subjective feedback information of user feedback corresponding to the multimedia resources, and acquiring a characteristic value of the user feedback information according to the quantization result of the objective feedback information and/or the subjective feedback information.
The obtaining the characteristic value of the user feedback information according to the quantization result of the objective feedback information and/or the subjective feedback information includes:
and obtaining the characteristic value of the user feedback information according to the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the objective feedback information and/or the subjective feedback information.
The obtaining the characteristic value of the user feedback information according to the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the objective feedback information and/or the subjective feedback information comprises the following steps:
and taking the result obtained after normalization processing of the coding result of the discrete feature and the derivative result of the continuous feature in the objective feedback information and/or the subjective feedback information as the feature value of the user feedback information.
In this embodiment, the obtained user feedback information includes objective feedback information and/or subjective feedback information fed back by the user, where the objective feedback information may be operation information of the user on the multimedia resource, for example, viewing duration, purchase order amount, and the like, and by using the objective feedback information fed back by the user, objective operation of the user on the multimedia resource may be considered on the basis of comprehensive evaluation of the multimedia resource by using subjective evaluation of the user and objective evaluation of the multimedia resource itself, so as to further improve evaluation accuracy and objectivity of the multimedia resource.
In addition, the determining the evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource includes:
obtaining the clustering category of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the feedback information of the user;
obtaining a class center value of the clustering class;
and determining an evaluation result of the multimedia resource by using the class center value.
In another embodiment, the determining the evaluation result of the multimedia resource according to the feature value of the information and the feature value of the user feedback information corresponding to the multimedia resource includes:
obtaining the clustering category of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the feedback information of the user;
obtaining a central value of the clustering class;
obtaining the characteristic value of the information of the multimedia resource and the entropy value of the characteristic value of the user feedback information;
and determining an evaluation result of the multimedia resource by using the central value and the entropy value.
In this embodiment, the central value of the cluster category is considered, and meanwhile, the entropy value of the feature value is considered, and the evaluation scale of the feature value on the multimedia resource is considered through the entropy value.
In addition, the embodiment of the disclosure also provides a method for pushing multimedia resources, which comprises the following steps:
acquiring a characteristic value of information of at least one multimedia resource in a preset time window;
determining an evaluation result of each multimedia resource according to the characteristic value of the information of each multimedia resource and the characteristic value of the user feedback information corresponding to each multimedia resource;
and pushing the multimedia resources according to the evaluation results of the multimedia resources.
In this embodiment, the method for obtaining the feature value of the information of at least one multimedia resource in the preset time window includes:
determining at least one multimedia resource of a preset time window from a database;
and acquiring personnel characteristic information and/or video characteristic information of each multimedia resource, and acquiring characteristic values of corresponding information according to the quantized result of the personnel characteristic information and/or the video characteristic information of each multimedia resource.
The method for obtaining the characteristic value of the corresponding information according to the personnel characteristic information and/or the quantization result of the video and audio characteristic information of each multimedia resource comprises the following steps:
and obtaining the characteristic value of the corresponding information according to the personnel characteristic information and/or the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the video and audio characteristic information of each multimedia resource.
Wherein, the obtaining the feature value of the corresponding information according to the encoding result of the discrete feature and the derivative result of the continuous feature in the personnel feature information and/or the video feature information of each multimedia resource comprises:
and taking the normalized results of the encoding results of the discrete features and the derived results of the continuous features in the personnel feature information and/or the video feature information of each multimedia resource as the feature values of the corresponding information.
The method for acquiring the characteristic value of the user feedback information corresponding to each multimedia resource comprises the following steps:
objective feedback information and/or subjective feedback information of user feedback corresponding to each multimedia resource are obtained,
and obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the quantization result of the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource.
The obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the quantization result of the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource includes:
and obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the objective feedback information corresponding to each multimedia resource and/or the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the subjective feedback information.
The obtaining the feature value of the user feedback information corresponding to each multimedia resource according to the encoding result of the discrete feature and the derivative result of the continuous feature in the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource includes:
and taking the results obtained after normalization processing of the coding results of the discrete features and the derivative results of the continuous features in the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource as the feature values of the user feedback information corresponding to each multimedia resource.
In addition, in this embodiment, after obtaining the feature value of the information of at least one multimedia resource and the feature value of the user feedback information corresponding to each multimedia resource, a feature value matrix may be constructed, where the lateral quantity of the feature value matrix is used to represent the feature value of the information of the same multimedia resource and the feature value of the corresponding user feedback information; or alternatively, the process may be performed,
the longitudinal quantity of the characteristic value matrix is used for representing the characteristic value of the information of the same multimedia resource and the characteristic value of the corresponding user feedback information.
In addition, in this embodiment, the determining the evaluation result of each multimedia resource according to the feature value of the information of each multimedia resource and the feature value of the user feedback information corresponding to each multimedia resource includes:
Inputting the eigenvalue matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
determining an evaluation result of each multimedia resource by utilizing various center values;
wherein each cluster category corresponds to one or more multimedia resources.
Or inputting the eigenvalue matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
obtaining a corresponding entropy value according to the characteristic value of the information of each multimedia resource and the characteristic value of the feedback information of the user;
and determining an evaluation result of each multimedia resource by using the class center value of each cluster class and the entropy value corresponding to each multimedia resource.
In addition, in this embodiment, the pushing the multimedia resources according to the evaluation result of each multimedia resource includes:
and pushing the multimedia resources to the user according to the sequencing result of the evaluation result of each multimedia resource.
As can be seen from the above embodiments, when pushing the multimedia resources, the present disclosure pushes the multimedia resources according to the sorting result of the evaluation results of each multimedia resource, where the evaluation results of each multimedia resource consider subjective evaluation of the user and at least consider objective evaluation of the multimedia resource, so that the evaluation result is more objective and accurate, and further the result of pushing each multimedia resource is also more objective and accurate.
According to another aspect of the present disclosure, at least one embodiment provides a server, including:
the acquisition module is used for acquiring the characteristic value of the information of the multimedia resource;
and the processing module is used for determining an evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource.
In addition, the acquisition module is further used for acquiring the characteristic value of the user feedback information corresponding to the multimedia resource.
Wherein the determining the evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource comprises:
obtaining the clustering category of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the feedback information of the user;
obtaining a class center value of the clustering class;
and determining an evaluation result of the multimedia resource by using the class center value.
Wherein the determining the evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource comprises:
Obtaining the clustering category of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the feedback information of the user;
obtaining a central value of the clustering class;
obtaining the characteristic value of the information of the multimedia resource and the entropy value of the characteristic value of the user feedback information;
and determining an evaluation result of the multimedia resource by using the central value and the entropy value.
According to another aspect of the present disclosure, at least one embodiment provides a server, including: a processor, a memory, and a program stored on the memory and executable on the processor, which when executed by the processor, performs the method steps in the embodiments described above.
According to another aspect of the present disclosure, at least one embodiment provides a server, including:
the acquisition module is used for acquiring the characteristic value of the information of at least one multimedia resource in the preset time window;
the processing module is used for determining an evaluation result of each multimedia resource according to the characteristic value of the information of each multimedia resource and the characteristic value of the user feedback information corresponding to each multimedia resource;
and the pushing module is used for pushing the multimedia resources according to the evaluation results of the multimedia resources.
In addition, the acquisition module is used for determining at least one multimedia resource of a preset time window from a database;
and acquiring personnel characteristic information and/or video characteristic information of each multimedia resource, and acquiring characteristic values of corresponding information according to the quantized result of the personnel characteristic information and/or the video characteristic information of each multimedia resource.
Wherein the acquisition module is also used for acquiring objective feedback information and/or subjective feedback information of user feedback corresponding to each multimedia resource,
and obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the quantization result of the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource.
The processing module is further used for constructing a characteristic value matrix by utilizing the characteristic value of the information of at least one multimedia resource and the characteristic value of the user feedback information corresponding to each multimedia resource;
the transverse quantity of the characteristic value matrix is used for representing characteristic values of information of the same multimedia resource and corresponding characteristic values of user feedback information; or alternatively, the process may be performed,
the longitudinal quantity of the characteristic value matrix is used for representing the characteristic value of the information of the same multimedia resource and the characteristic value of the corresponding user feedback information.
Wherein the processing module is used for processing the data,
the characteristic value matrix is used for inputting the characteristic value matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
determining an evaluation result of each multimedia resource by utilizing various center values;
wherein each cluster category corresponds to one or more multimedia resources.
In addition, the processing module is used for processing the data,
the characteristic value matrix is used for inputting the characteristic value matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
obtaining a corresponding entropy value according to the characteristic value of the information of each multimedia resource and the characteristic value of the feedback information of the user;
and determining an evaluation result of each multimedia resource by using the class center value of each cluster class and the entropy value corresponding to each multimedia resource.
In addition, the pushing module is used for pushing the multimedia resources to the user according to the sorting result of the evaluation result of each multimedia resource.
According to another aspect of the present disclosure, at least one embodiment provides a server, including: a processor, a memory and a program stored on the memory and executable on the processor, which when executed by the processor, implements method steps in pushing multimedia resources.
According to another aspect of the present disclosure, at least one embodiment provides a computer-readable storage medium having a program stored thereon, which when executed by a processor, implements the method steps of the embodiments described above.
The embodiment of the invention utilizes the information of the multimedia resource as one of the conditions for forming the evaluation result of the multimedia resource, can realize comprehensive evaluation of the multimedia resource by utilizing the subjective evaluation of the user and the objective evaluation of the multimedia resource, and ensures that the evaluation result of the multimedia resource is more objective and accurate.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
fig. 1 is a flow chart of a method for evaluating multimedia resources according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating another method for evaluating multimedia resources according to an embodiment of the present invention;
Fig. 3 is a flowchart of another method for pushing multimedia resources according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a server provided by an embodiment of the present invention;
fig. 5 is a schematic structural diagram of another service end according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of another service end according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The terms first, second and the like in the description and in the claims, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be capable of operation in sequences other than those illustrated or described herein, for example. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. "and/or" in the specification and claims means at least one of the connected objects.
The following description provides examples and does not limit the scope, applicability, or configuration as set forth in the claims. Changes may be made in the function and arrangement of elements discussed without departing from the spirit and scope of the disclosure. Various examples may omit, substitute, or add various procedures or components as appropriate. For example, the described methods may be performed in an order different than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
With the development of society, people need not only support in terms of substances but also mental supply in daily life production, especially multimedia resources such as movies, television shows, variety programs, music, dances, and the like. The multimedia resource is a popular and easily understood culture with strong comprehensiveness, on one hand, people of different age groups can obtain thought content which the author wants to express for the user more intuitively through the multimedia resource, and on the other hand, people with excessive mental stress can be relieved through the multimedia resource, so that the stability of society is stabilized.
However, since the output of the multimedia resources is affected by the author, actor, director, etc., there is a distinction between the quality and the quality of the produced multimedia resources, and in order to enable the user to more contact the high-quality multimedia resources, many multimedia resource playing software provides a scoring system.
For example, the bean cotyledon score is used as a marker post for evaluating the quality of movies and videos, and provides quantization indexes of multimedia resources such as different movies, television shows, and various arts. Its evaluation is mainly derived from the mass evaluation group. The registered user of the bean cotyledon can score one to five stars each time he finishes watching one multimedia resource. For example, a movie has 10 ten thousand users scored by converting the star scale of 10 ten thousand users into 0 to 10 points, and then calculating the score average of 10 ten thousand users as the evaluation value of the movie. And then, at intervals, automatically updating and adding the score of the new user, thereby quickly showing timely evaluation. Each user is made by 'one-person-one-ticket', along with the increase of the number of scoring persons, the influence of single person scoring on the overall scoring trend is continuously diluted, and the bean cotyledon scoring is still used for evaluating the multimedia resources from the subjective angle of the user although a plurality of evasions and precautionary measures are provided, so that the evaluation result lacks objectivity due to the fact that the multidimensional behavior characteristics are not integrated.
Likewise, third party mainstream platforms such as Tencen, aiqi, you cool and the like also have own scoring mechanisms, and Tencen determines the evaluation result of the multimedia resource by integrating the popularity, the comment number, the praise number and the bean score of the media resource; the evaluation result of the multimedia resource is determined by comprehensively calculating the scoring and top stepping amounts of the platform users, and the evaluation result is independent and relatively independent without referring to the playing amounts and bean scores. The preference is to comprehensively determine the evaluation result of the multimedia resources by referring to the factors such as heat, freshness, consumption trend and the like, and it can be seen that the favorite superior playing platform evaluates the multimedia resources through the scores of users. The multimedia resources are evaluated from a multidimensional point of view without quantifying the behavioral characteristics of the user.
From the above, it can be seen that the current evaluation mode of the multimedia resources by each network platform is subjective evaluation after the user views, and the evaluation effect mainly depends on the mood of the user and lacks objectivity.
In order to solve the technical problems presented above, the embodiments of the present disclosure provide a method for evaluating and pushing a multimedia resource and a server, which can implement comprehensive evaluation of the multimedia resource from multiple dimensions such as the multimedia resource itself and feedback information of a user when evaluating the multimedia resource, so as to obtain objective and accurate evaluation results.
Referring to fig. 1, an embodiment of the disclosure provides a method for evaluating a multimedia resource, including:
s101, acquiring characteristic values of information of multimedia resources;
s102, determining an evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource.
In this embodiment, the multimedia is an integration of various media, and generally includes multimedia forms such as text, sound, and image.
In a computer system, multimedia refers to a man-machine interactive information communication and media combining two or more media. The media used include text, pictures, photographs, sounds, animations and movies, and interactive functions provided by the program.
Thus, the multimedia assets include text assets, sound assets, picture assets, photo assets, sound assets, animation assets, movie assets, and the like.
In step S101, the information of the multimedia resource may include: personnel characteristic information and/or video characteristic information. Wherein the personal characteristic information may comprise personal characteristic information related to the multimedia asset itself, the personal characteristic information comprising, in one illustrative example: director characteristic information and actor characteristic information.
The director characteristic information may include personal characteristic information of a director, a characteristic number of winning a video and audio that the director has conducted in the past (for example, berlin international movie festival bear prize, canne movie festival palm prize, etc.), and box office characteristic information of a video and audio that the director has conducted, etc.
The actor characteristic information may illustratively include information of a lead actor characteristic information, a popularity of the lead actor, a professional index, a speaking index, and the like. The popularity of the actor may include the influence degree of a platform with public influence such as hundred degrees, microblogs, etc. (the number of active fans of the actor, the number of interactions with the fans, the influence on the platform, the search index of the platform, etc.). Wherein, the professional index of each actor mainly refers to the winning number of other works of the main actor, the good score of the performance, and the like. The expression index refers to the brand level of the actors and the brand number of the expression.
The audio-visual characteristic information may include, for example, a tag, a time of showing, an update time, a language type, a payment type, etc. of the audio-visual characteristic information.
In practical applications, the information of the multimedia resource may include personnel feature information and audio-visual feature information of the multimedia resource, or one of the personnel feature information and the audio-visual feature information may be selected, and in particular, according to the evaluation scale of the multimedia resource, generally, the more the information of the multimedia resource is, the higher the objective evaluation scale of the multimedia resource is, and the more objective the evaluation result of the multimedia resource is. Otherwise, the less the information of the multimedia resource is, the lower the objective evaluation scale of the multimedia resource is, and the evaluation result of the multimedia resource is biased to subjective.
In one embodiment, it is assumed that the information of the multimedia resource includes personnel feature information and audio-visual feature information, and at the same time, the personnel feature information and the audio-visual feature information are as described in the above examples, the acquired information may be recorded in a functional manner, as shown in the following examples:
F 1 ＝[f d ,f a ,f m ]
wherein F is 1 Information for representing multimedia resources, f d Representing director characteristic information, each dimension characteristic information related to the director can be recorded as f d ＝[f d1 ,f d2 ...f dd ]D represents the number of director features; wherein f a Representing personnel characteristic information, each dimension characteristic information related to personnel can be recorded as f a ＝[f a1 ,f a2 ...f aa ]A represents the characteristic number of the actor; wherein f m Representing video and audio feature information, and each dimension feature information related to video and audio can be recorded as f m ＝[f m1 ,f m2 ...f mm ]M represents the number of film features.
In step S102, in order to obtain an accurate evaluation result, the embodiment uses the obtained characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource to perform multi-dimensional evaluation on the multimedia resource.
In this embodiment, the user feedback information includes objective feedback information and/or subjective feedback information of user feedback, where the objective feedback information includes an operation action instruction of the user for the multimedia resource, for example, a number of users watching the multimedia resource, a number of users collecting the multimedia resource, a number of users searching the multimedia resource, a number of users subscribing to the multimedia resource, and so on.
In one embodiment, the objective feedback information of the user feedback obtained may be recorded as a function, as shown in the following example:
F 2 ＝[f p ,f f ,f s ,f o ]
Wherein F is 2 Objective feedback information for representing user feedback, f p Feature for representing the number of users viewing the multimedia asset, f f For representing the number of users who collect the multimedia resource, f s Feature for indicating the number of users searching for the multimedia resource, f o For representing the number of users subscribed to the multimedia asset.
In practical application, obtain f p In this case, the number of watched multimedia resources (for example, when the multimedia resources are divided into a plurality of episodes, the number of watched episodes is obtained, the number of watched multimedia resources is obtained, and the number of recommended multimedia resources is obtained), the average time length of watched multimedia resources is long, and in order to avoid that some multimedia resources are updated only once a week, the watching amount of the current day is large, the watching amount of other time is small, and the situation that the multimedia resources belonging to high quality are mispredicted as non-high quality multimedia resources occurs, and the watching number of the updated multimedia resources and the watching times of the updated multimedia resources can be introduced.
Acquiring f f And when the user is in the process of collecting the multimedia resources, the number of people and the number of times the multimedia resources are collected are obtained.
Acquiring f s And acquiring the number of people with the searched multimedia resources and the number of times the multimedia resources are searched.
Acquiring f o Acquiring revenue obtained from and ordering the multimedia assetOrder quantity of source.
In one embodiment, the subjective feedback information of the user feedback obtained may be recorded as a function, as shown in the following example:
F 3 ＝f q
wherein f q For representing subjective evaluations of multimedia assets by a user.
In practical application, f q The user rating of the multimedia resource may be obtained from the multimedia resource playing software carrying the multimedia resource, for example, from the multimedia resource playing software such as a bean-shaped video, a cool video, a Tencent video, an aiqi screen, etc. (in one case, the user rating of the multimedia resource may be obtained from any of the above-mentioned multimedia resource playing software, and the obtained user rating may be used as the user rating of the multimedia resource.
F of the multimedia resource is obtained by the method 1 、F 2 、F 3 After that, can be used for F 1 、F 2 、F 3 Feature stitching is performed, and the stitching is a feature function F= [ F ] 1 ,F 2 ,F 2 ]＝[f d ,f a ,f m ,f p ,f f ,f s ,f o ,f q ]。
It should be noted that, the characteristic value of the information of the multimedia resource and the characteristic value of the corresponding user feedback information may be variable, for example, the characteristic function f= [ F ] 1 ,F 2 ,F 2 ]＝[f d ,f a ,f m ,f p ,f f ,f s ,f o ,f q ]May also be F= [ F ] 1 ,F 2 ,F 2 ]＝[f d ,f a ,f p ,f f ,f s ,f o ,f q ]Or may alsoTo be F= [ F ] 1 ,F 2 ,F 2 ]＝[f d ,,f m ,f p ,f s ,f o ,f q ]The feature values and feature functions provided in the present embodiment are only exemplary, and the types and types of the specific feature values and feature functions are not limited.
It should be noted that, the characteristic value of the information of the multimedia resource acquired in step S101 may be at least one information characteristic of the multimedia resource (one or more characteristic information of the multimedia resource) in a preset time window, and the preset time window may be set according to the requirement of the administrator, for example, set to be a week, a month, a year, etc.
As can be seen from the above-mentioned process of acquiring the characteristics of the information and the characteristics of the corresponding user feedback information, when the characteristic information is acquired, the acquired characteristic information includes discrete characteristics and continuous characteristics, for example, taking the acquired director characteristic information as an example, the popularity of the director, the film name of the director, and the like are discrete characteristics, and the number of times of winning a prize in a period of the director is continuous characteristics. Thus, in order to generate F 1 The acquired information needs to be quantized, which includes coding discrete feature information (e.g., label encoding), deriving continuous feature information, and obtaining quantization results.
However, since the dimension of each dimension characteristic information of the multimedia resource is different, for example, f p Duration and f of the multimedia asset being viewed by the user o The order quantity of the multimedia resource in the system does not belong to the same dimension, so that in order to eliminate the influence of the dimension, the feature information of each dimension can be expressed under the same dimension, and normalization processing is required for the obtained feature information of each dimension.
For example, the characteristic information of the information of the multimedia resource and the corresponding characteristic of the user feedback information are subjected to quantization processing and normalization processing.
Specifically, the embodiment of the disclosure exemplarily provides a formula of normalization processing, which is as follows:
F der ＝Derive(F)
Figure BDA0003352273500000141
wherein F is the characteristic function obtained in the above embodiment, D= { D, a, m, p, F, s, p, q }, F imin Is the minimum value of the sample data, f imax Is the maximum value of the sample data.
And obtaining the normalized characteristic function through the formula.
When step S102 is performed, the obtained feature function F (because the features of the information of the at least one multimedia resource that may be the preset time window are described above, and thus, F may be a feature value matrix here) is input into a clustering model (for convenience of explanation, a spectral clustering model is taken as an example later) to obtain a clustering class, and an evaluation result of the multimedia resource is determined according to a class center value of each clustering class.
Specifically, the spectral clustering model can classify the categories according to the requirements, if the multimedia resources are scored by 1-10 minutes, the category numbers can be classified into 10 categories, and after F is input into the spectral clustering model, the multimedia resources can be classified into different category clusters by gathering the same characteristics into one category.
The spectrum clustering divides the weighted undirected graph into a plurality of optimal subgraphs, so that the internal features of the subgraphs are similar as much as possible, the distance between the subgraphs is as far as possible, the purpose of clustering is achieved, and the embodiment provides a spectrum clustering formula in an exemplary way, as follows:
M sc ＝{(m i ,k)|k,i∈N + ,1≤k≤10,1≤i≤n}
Figure BDA0003352273500000142
wherein M is sc Representing the spectral clustering result of each multimedia resource, k is the clustering category of the spectral clustering, m i Representing the ith multimedia asset.
After the feature functions are clustered, the average value of the feature values of different dimensions of the multimedia resources in each class is calculated to serve as the feature value of the class center of the class, the feature of the class center of the k class is averaged, namely the average value of the multi-dimensional feature of each class center is calculated to serve as an evaluation mark of the class, and then the evaluation mark is compared to serve as the scoring of different classes. So that the value scores of the multimedia resources in the same class are the same (the value scores are the evaluation results of the multimedia resources in the class), for example, 10 scores are given to the multimedia resources in 10 classes.
In this embodiment, a class center evaluation model is exemplarily provided for implementing the above calculation of a class center value, where the class center evaluation model includes:
Figure BDA0003352273500000151
centerk∈[1,2..10],sc(i)＝centerk,i∈n
score cluster (m i )＝score cluster (m centerk )＝mean(f dcenteri ,f acenteri ,f mcenteri ,f pcenteri ,f fcenteri ,f scenteri ,f ocenteri ,f qcenteri ,)
wherein m is centerk And (3) representing the class center after the multimedia resources are clustered, namely the average evaluation level of each multimedia resource belonging to the class.
score cluster (m centerk ) And the score of the class center is represented, wherein the score value is the average value of all dimension characteristics of the multimedia resources, namely, the score of each multimedia resource belonging to the class is equal to the score of the class center.
score cluster (m i ) And after the clustering, the score of each multimedia resource in the class is represented, and the score can be used as an evaluation result of each multimedia resource.
According to the embodiments, the method for evaluating the multimedia resources provided by the disclosure can perform multi-dimensional comprehensive evaluation according to the objective dimension of the multimedia resources and in combination with the corresponding user feedback dimension (including objective information and/or subjective information fed back by the user), so that the evaluation result of the multimedia resources is prevented from being too dependent on subjective evaluation of the user, and the objectivity and accuracy of the evaluation result of the multimedia resources are effectively improved.
In addition, in order to determine the influence relationship of the evaluation of each feature dimension on the evaluation result when the multimedia resource is evaluated by the above method in performing step S102, in another embodiment provided in the present disclosure, an entropy value is introduced in this embodiment, and in particular in performing step S102, in addition to obtaining the clustering class of the multimedia resource and obtaining the central value of the clustering class according to the feature value of the information of the multimedia resource and the feature value of the user feedback information, the entropy value of each dimension feature (the feature value of the information and the feature value of the user feedback information) of the multimedia resource is obtained, and the evaluation result of the multimedia resource is determined by using the central value of the class and the entropy value.
To fully illustrate the aspects of the present disclosure, the present disclosure illustratively provides a method for entropy calculation, as follows:
acquiring dimensional characteristics of multimedia resources, i.e. in the above embodiments
F n ＝[f dn ,f an ,f mn ,f pn ,f fn ,f sn ,f on ,f qn ]
Where n represents the number of multimedia resources.
And carrying out normalization processing on the acquired specific dimension characteristics, wherein the formula is as follows:
forward index:
Figure BDA0003352273500000161
negative index:
Figure BDA0003352273500000162
wherein x is ij The j-th feature of the i-th multimedia asset is represented.
Meanwhile, calculating the proportion of the ith multimedia resource in the index under the jth characteristic:
Figure BDA0003352273500000163
the entropy value of the j-th feature is calculated,
e j ≥0
Figure BDA0003352273500000164
calculating information entropy redundancy:
d j ＝1-e j
calculating the weight of each dimension feature:
Figure BDA0003352273500000165
calculating the score of each multimedia resource, and mapping into an interval of 1-10 partitions:
Figure BDA0003352273500000166
and obtaining the entropy value of the multimedia resource through the entropy value model, then utilizing model fusion, and finally obtaining the evaluation result of the multimedia resource according to the class center value and the entropy value.
The model fusion comprises scoring obtained by clustering and scoring weighted fusion obtained by an entropy method, and the fused result is used as an evaluation result of the multimedia resource.
In this embodiment, the entropy value of each dimension feature of the multimedia resource is introduced and combined with the class center value of the multimedia resource to determine the evaluation result of the multimedia resource, and the degree of dispersion of the multi-dimension feature can be determined by the entropy value to evaluate the influence of the feature on the evaluation result. For example, the smaller the entropy value, the larger the information amount, the larger the degree of dispersion, and the larger the degree of influence of the feature (one of the feature of the information and the feature of the user feedback information) on the evaluation result of the multimedia resource, the larger the feature weight. The smaller the entropy value of the same theory, the smaller the influence degree of the feature on the evaluation result of the multimedia resource, and the smaller the feature weight.
By the embodiment, the method and the device comprehensively consider the multidimensional features of the multimedia resources, conduct feature derivation and feature fusion on the information features of the multimedia resources, the objective features of user feedback and the subjective features of user feedback, increase a class center evaluation center module, conduct modeling prediction evaluation on the multidimensional features of the multimedia resources, measure the influence degree of the features on evaluation scores from the single-dimensional features by using an entropy method, enrich a multimedia resource value evaluation system and distinguish high-quality multimedia resources from non-high-quality multimedia resources.
To fully illustrate the above embodiments, the present disclosure also provides the following embodiments, as shown in fig. 2, in combination with actual data. (the following embodiments are described in connection with actual data, and the quantization process, clustering process, class center value calculation process, and entropy value calculation process for each feature are described in the foregoing embodiments, and repetitive description will not be made later.)
S201, determining a preset time window.
S202, acquiring characteristic values for evaluating each multimedia resource, wherein the characteristic values comprise: the characteristic value of the information, objective feedback information and subjective feedback information fed back by a user.
Wherein the information includes: director characteristics, actor characteristics, film characteristics, and objective feedback information fed back by the user comprises: the method comprises the steps of taking bean scores of multimedia resources as subjective characteristics fed back by a user, wherein the subjective characteristics comprise viewing characteristics of the user for viewing the multimedia resources, collecting characteristics of the user for collecting the multimedia resources, searching characteristics of the user for searching the multimedia resources, ordering characteristics of the user for ordering the multimedia resources and the like.
S203, after each feature acquired in the step S202 is quantized, a feature function is acquired
Figure BDA0003352273500000171
Will be quantized
Figure BDA0003352273500000181
The normalized spectrum clustering model is input, and the obtained clustering result is assumed to be { m } 1 ,m 2 One kind, { m } 4 One kind, { m } 3 ,m 5 One type is }, where m 1 、m 2 、m 3 、m 4 、m 5 Representing different multimedia resources in F.
In step S204, class centers of the classes are calculated:
Figure BDA0003352273500000182
calculating the central values of various centers:
Figure BDA0003352273500000183
if score is center1 >score center2 >score center3 Then score (m 1 )＝score(m 2 )＝3，score(m 4 )＝2，
score(m 2 )＝score(m 5 )＝1
At this time, the class center value of each class can be used as the evaluation result of each multimedia resource in the class through the calculated class center value of each class.
As in the foregoing embodiments, in order to consider the influence of each feature of the multimedia resource on the evaluation result, an entropy value may also be introduced, as in step S205.
In step S205, for
Figure BDA0003352273500000184
Entropy calculation is carried out on the characteristics of each multimedia resource in the system to obtain a multimedia resource m 1 For example, if m is calculated 1 Is given by
w 1 ＝0.1，w 2 ＝0.1，w 3 ＝0.05，w 4 ＝0.05，w 5 ＝0.2，w 6 ＝0.2，w 7 ＝0.1，w 8 ＝0.2
Then m 1 The corresponding entropy value is score entropy_m1 ＝w1*p11+w2*p12...+w8*p18＝4
In step S206, the final evaluation result is calculated as 3.5 by combining the class center value and the entropy value.
Figure BDA0003352273500000185
According to the embodiment, the multi-dimensional characteristics of the multimedia resources are comprehensively considered, the information characteristics, the objective characteristics of user feedback and the subjective characteristics of user feedback of the multimedia resources are subjected to characteristic derivation and characteristic fusion, a class center evaluation center module is added, modeling, prediction and evaluation are carried out on the multi-dimensional characteristics of the multimedia resources, the influence degree of the characteristics on evaluation scores is measured from the single-dimensional characteristics by utilizing an entropy method, a multimedia resource value evaluation system is enriched, and high-quality multimedia resources and non-high-quality multimedia resources are distinguished.
In addition, in combination with the above embodiments of the method for evaluating multimedia resources, the embodiments of the present disclosure further provide a method for pushing multimedia resources, as shown in fig. 3, where the method includes:
s301, obtaining a characteristic value of information of at least one multimedia resource in a preset time window;
s302, determining an evaluation result of each multimedia resource according to the characteristic value of the information of each multimedia resource and the characteristic value of the user feedback information corresponding to each multimedia resource;
S303, pushing the multimedia resources according to the evaluation results of the multimedia resources.
In step S301, at least one multimedia resource of a preset time window is determined from the database, where the preset time window may be set to one week, and the multimedia resource includes a text resource, a picture resource, a photo resource, an audio-visual resource, and the like.
And acquiring personnel characteristic information and/or video characteristic information of each multimedia resource, and acquiring characteristic values of corresponding information according to the quantized result of the personnel characteristic information and/or the video characteristic information of each multimedia resource.
The method for obtaining the characteristic value of the corresponding information according to the personnel characteristic information and/or the quantization result of the video and audio characteristic information of each multimedia resource comprises the following steps:
and obtaining the characteristic value of the corresponding information according to the personnel characteristic information and/or the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the video and audio characteristic information of each multimedia resource.
In order to eliminate the influence of dimension, the method for obtaining the feature value of the corresponding information according to the encoding result of the discrete feature and the derivative result of the continuous feature in the personnel feature information and/or the video feature information of each multimedia resource comprises the following steps:
And taking the normalized results of the encoding results of the discrete features and the derived results of the continuous features in the personnel feature information and/or the video feature information of each multimedia resource as the feature values of the corresponding information.
When executing step S302, obtaining the user feedback information corresponding to each multimedia, including objective feedback information and subjective feedback information of the user feedback, and obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the quantization result of the objective feedback information and/or subjective feedback information corresponding to each multimedia resource.
The obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the quantization result of the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource includes:
and obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the objective feedback information corresponding to each multimedia resource and/or the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the subjective feedback information.
Also, to avoid the influence of dimension, the obtaining the feature value of the user feedback information corresponding to each multimedia resource according to the encoding result of the discrete feature and the derivative result of the continuous feature in the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource includes:
And taking the results obtained after normalization processing of the coding results of the discrete features and the derivative results of the continuous features in the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource as the feature values of the user feedback information corresponding to each multimedia resource.
In this embodiment, a feature value matrix is constructed by using feature values of information of at least one multimedia resource and feature values of user feedback information corresponding to each multimedia resource (i.e., F in the above embodiment), where a lateral amount of the feature value matrix is used to represent feature values of information of the same multimedia resource and feature values of corresponding user feedback information; or alternatively, the process may be performed,
the longitudinal quantity of the characteristic value matrix is used for representing the characteristic value of the information of the same multimedia resource and the characteristic value of the corresponding user feedback information.
When executing step S302, determining an evaluation result of each multimedia resource according to the feature value of the information of each multimedia resource and the feature value of the user feedback information corresponding to each multimedia resource, including:
inputting the eigenvalue matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
Determining an evaluation result of each multimedia resource by utilizing various center values;
wherein each cluster category corresponds to one or more multimedia resources.
In this embodiment, in order to determine the influence of each dimension feature on the evaluation result, the entropy value of the feature may also be introduced.
After the entropy value is introduced, the evaluation result of the multimedia resource is determined by the following method.
Inputting the eigenvalue matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
obtaining a corresponding entropy value according to the characteristic value of the information of each multimedia resource and the characteristic value of the feedback information of the user;
and determining an evaluation result of each multimedia resource by using the class center value of each cluster class and the entropy value corresponding to each multimedia resource.
In step S303, each multimedia resource is pushed according to a preset rule according to the evaluation result of each multimedia resource, where the preset rule may include pushing the multimedia resource to the user according to the sorting result of the evaluation result of each multimedia resource.
According to the embodiments, when the multimedia resource software (database) is in the pushing resource library, the evaluation result of each multimedia resource can be obtained, the multimedia resources are ordered according to the evaluation result of the multimedia resources, and the pushing result of the multimedia resources pushed to the user is more objective and accurate.
Based on the above method embodiments, the embodiment of the present disclosure further provides a server, as shown in fig. 4, where the server includes: acquisition module 401, processing module 402
An acquisition module 401, configured to acquire a feature value of information of a multimedia resource;
and the processing module 402 is configured to determine an evaluation result of the multimedia resource according to the feature value of the information and the feature value of the user feedback information corresponding to the multimedia resource.
Wherein, this server side includes: server, PC side, mobile terminal, etc.
In this embodiment, the obtaining module 401 is further configured to obtain a feature value of user feedback information corresponding to the multimedia resource.
The user feedback information comprises objective information and/or subjective information fed back by the user.
The processing module 402 obtains the clustering category of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the user feedback information;
obtaining a class center value of the clustering class;
and determining an evaluation result of the multimedia resource by using the class center value.
In other embodiments, the processing module 402 obtains the clustering type of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the user feedback information;
Obtaining a central value of the clustering class;
obtaining the characteristic value of the information of the multimedia resource and the entropy value of the characteristic value of the user feedback information;
and determining an evaluation result of the multimedia resource by using the central value and the entropy value.
According to the embodiment, the server can determine the evaluation result of the multimedia resource according to each dimension characteristic of the multimedia resource, so that the technical problem that the evaluation result is excessively influenced by subjective evaluation and inaccurate in evaluation of the multimedia resource caused by taking subjective evaluation of a user as the evaluation result of the multimedia resource in the prior art is avoided.
The apparatus in this embodiment is a device corresponding to the method shown in fig. 1, and the implementation manner in each embodiment is applicable to the embodiment of the device, so that the same technical effects can be achieved. It should be noted that, the above device provided in the embodiment of the present invention can implement all the method steps implemented in the method embodiment and achieve the same technical effects, and detailed descriptions of the same parts and beneficial effects as those of the method embodiment in the embodiment are omitted herein.
The embodiment of the disclosure also provides a server, which comprises: a processor, a memory, and a program stored on the memory and executable on the processor, which when executed by the processor, implements the embodiments related to fig. 1.
The embodiment of the disclosure further provides a server, as shown in fig. 5, where the server includes:
an acquisition module 501, a processing module 502, a pushing module 503,
an obtaining module 501, configured to obtain a feature value of information of at least one multimedia resource in a preset time window;
the processing module 502 is configured to determine an evaluation result of each multimedia resource according to the feature value of the information of each multimedia resource and the feature value of the user feedback information corresponding to each multimedia resource;
and the pushing module 503 is configured to push the multimedia resources according to the evaluation results of the multimedia resources.
The obtaining module 501 is configured to obtain personnel feature information and/or audio-visual feature information of each multimedia resource, and obtain a feature value of the corresponding information according to a quantization result of the personnel feature information and/or the audio-visual feature information of each multimedia resource.
The obtaining module 501 is further configured to obtain objective feedback information and/or subjective feedback information of user feedback corresponding to each multimedia resource,
and obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the quantization result of the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource.
The processing module 502 is further configured to construct a feature value matrix by using feature values of information of at least one multimedia resource and feature values of user feedback information corresponding to each multimedia resource;
the transverse quantity of the characteristic value matrix is used for representing characteristic values of information of the same multimedia resource and corresponding characteristic values of user feedback information; or alternatively, the process may be performed,
the longitudinal quantity of the characteristic value matrix is used for representing the characteristic value of the information of the same multimedia resource and the characteristic value of the corresponding user feedback information
The processing module 502 is specifically configured to input the eigenvalue matrix into a clustering model, and obtain a clustering class output by the clustering model;
obtaining class center values of all cluster classes;
determining an evaluation result of each multimedia resource by utilizing various center values;
wherein each cluster category corresponds to one or more multimedia resources.
In other embodiments, the processing module 502 is further configured to input the feature value matrix into a cluster model, and obtain a cluster class output by the cluster model;
obtaining class center values of all cluster classes;
obtaining a corresponding entropy value according to the characteristic value of the information of each multimedia resource and the characteristic value of the feedback information of the user;
And determining an evaluation result of each multimedia resource by using the class center value of each cluster class and the entropy value corresponding to each multimedia resource.
The pushing module 503 is specifically configured to push the multimedia resources to the user according to the sorting result of the evaluation result of each multimedia resource.
The apparatus in this embodiment is a device corresponding to the method shown in fig. 3, and the implementation manner in each embodiment is applicable to the embodiment of the device, so that the same technical effects can be achieved. It should be noted that, the above device provided in the embodiment of the present invention can implement all the method steps implemented in the method embodiment and achieve the same technical effects, and detailed descriptions of the same parts and beneficial effects as those of the method embodiment in the embodiment are omitted herein.
As shown in fig. 6, the embodiment of the present disclosure further provides a server 60, including: a processor 61, a memory 62 and a program stored on the memory 62 and executable on the processor, which when executed by the processor, implements the embodiments related to the above methods.
The presently disclosed embodiments also provide a computer-readable storage medium having stored thereon a program that, when executed by a processor, implements steps S101-S102 and steps S301-S303.
It can be appreciated that in the embodiment of the present invention, when the computer program is executed by the processor, the processes of the method embodiments shown in fig. 1 and fig. 3 can be implemented, and the same technical effects can be achieved, so that repetition is avoided, and no further description is provided herein.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown 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 units may be selected according to actual needs to achieve the purpose of the embodiment of the present invention.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising at least one instruction to cause a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, etc.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.

Claims (35)

1. A method of evaluating a multimedia asset, the method comprising:
acquiring a characteristic value of information of the multimedia resource;
and determining an evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource.
2. The method of claim 1, wherein the obtaining the characteristic value of the information of the multimedia resource comprises:
and acquiring the characteristic value of the information of at least one multimedia resource in the preset time window.
3. The method of claim 1, wherein the method for obtaining the characteristic value of the information of the multimedia resource comprises:
and acquiring personnel characteristic information and/or video characteristic information of the multimedia resource, and acquiring characteristic values of the information according to quantization results of the personnel characteristic information and/or the video characteristic information.
4. A method according to claim 3, wherein said obtaining the feature value of the information based on the quantization result of the personnel feature information and/or the video and audio feature information comprises:
and obtaining the characteristic value of the information according to the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the personnel characteristic information and/or the video characteristic information.
5. The method according to claim 4, wherein the obtaining the feature value of the information based on the encoding result of the discrete feature and the deriving result of the continuous feature in the personnel feature information and/or the video feature information comprises:
and normalizing the coding result of the discrete features and the derivative result of the continuous features in the personnel feature information and/or the video and audio feature information to obtain the feature value of the information.
6. The method according to claim 1, wherein the characteristic value of the user feedback information corresponding to the multimedia resource is obtained by:
objective feedback information and/or subjective feedback information of user feedback corresponding to the multimedia resources are obtained, and characteristic values of the user feedback information corresponding to the multimedia resources are obtained according to quantization results of the objective feedback information and/or the subjective feedback information.
7. The method according to claim 6, wherein the obtaining the characteristic value of the user feedback information according to the quantization result of the objective feedback information and/or the subjective feedback information includes:
and obtaining the characteristic value of the user feedback information according to the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the objective feedback information and/or the subjective feedback information.
8. The method according to claim 7, wherein the obtaining the feature value of the user feedback information according to the encoding result of the discrete feature and the derivative result of the continuous feature in the objective feedback information and/or the subjective feedback information comprises:
and taking the result obtained after normalization processing of the coding result of the discrete feature and the derivative result of the continuous feature in the objective feedback information and/or the subjective feedback information as the feature value of the user feedback information.
9. The method according to claim 1, wherein determining the evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource comprises:
obtaining the clustering category of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the feedback information of the user;
Obtaining a class center value of the clustering class;
and determining an evaluation result of the multimedia resource by using the class center value.
10. The method according to claim 1, wherein determining the evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource comprises:
obtaining the clustering category of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the feedback information of the user;
obtaining a central value of the clustering class;
obtaining the characteristic value of the information of the multimedia resource and the entropy value of the characteristic value of the user feedback information;
and determining an evaluation result of the multimedia resource by using the central value and the entropy value.
11. A method of pushing multimedia assets, the method comprising:
acquiring a characteristic value of information of at least one multimedia resource in a preset time window;
determining an evaluation result of each multimedia resource according to the characteristic value of the information of each multimedia resource and the characteristic value of the user feedback information corresponding to each multimedia resource;
and pushing the multimedia resources according to the evaluation results of the multimedia resources.
12. The method of claim 11, wherein the step of obtaining the characteristic value of the information of the at least one multimedia resource in the predetermined time window comprises:
determining at least one multimedia resource of a preset time window from a database;
and acquiring personnel characteristic information and/or video characteristic information of each multimedia resource, and acquiring characteristic values of corresponding information according to the quantized result of the personnel characteristic information and/or video characteristic information of each multimedia resource.
13. The method according to claim 12, wherein obtaining the feature value of the corresponding information according to the quantized result of the personnel feature information and/or the audio-visual feature information of each multimedia resource comprises:
and obtaining the characteristic value of the corresponding information according to the personnel characteristic information and/or the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the video and audio characteristic information of each multimedia resource.
14. The method according to claim 13, wherein the obtaining the feature value of the corresponding information according to the encoding result of the discrete feature and the deriving result of the continuous feature in the personnel feature information and/or the audio-visual feature information of each multimedia resource comprises:
And normalizing the coding result of the discrete feature and the derivative result of the continuous feature in the personnel feature information and/or the video feature information of each multimedia resource to obtain a feature value of the corresponding information.
15. The method according to claim 11, wherein the characteristic value of the user feedback information corresponding to the multimedia resource is obtained by:
objective feedback information and/or subjective feedback information of user feedback corresponding to each multimedia resource are obtained, and characteristic values of the user feedback information corresponding to each multimedia resource are obtained according to quantization results of the objective feedback information and/or subjective feedback information corresponding to each multimedia resource.
16. The method according to claim 15, wherein the obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the quantization result of the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource comprises:
and obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the coding result of the discrete characteristic and the derivative result of the continuous characteristic in the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource.
17. The method according to claim 16, wherein the obtaining the feature value of the user feedback information corresponding to each multimedia resource according to the encoding result of the discrete feature and the derivative result of the continuous feature in the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource includes:
and taking the results obtained after normalization processing of the coding results of the discrete features and the derivative results of the continuous features in the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource as the feature values of the user feedback information corresponding to each multimedia resource.
18. The method of claim 11, wherein the method further comprises:
constructing a characteristic value matrix by utilizing the characteristic value of the information of at least one multimedia resource and the characteristic value of the user feedback information corresponding to each multimedia resource;
the transverse quantity of the characteristic value matrix is used for representing characteristic values of information of the same multimedia resource and corresponding characteristic values of user feedback information; or alternatively, the process may be performed,
the longitudinal quantity of the characteristic value matrix is used for representing the characteristic value of the information of the same multimedia resource and the characteristic value of the corresponding user feedback information.
19. The method of claim 18, wherein determining the evaluation result of each multimedia resource according to the characteristic value of the information of each multimedia resource and the characteristic value of the user feedback information corresponding to each multimedia resource comprises:
inputting the eigenvalue matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
determining an evaluation result of each multimedia resource by utilizing various center values;
wherein each cluster category corresponds to one or more multimedia resources.
20. The method of claim 18, wherein determining the evaluation result of each multimedia resource according to the characteristic value of the information of each multimedia resource and the characteristic value of the user feedback information corresponding to each multimedia resource comprises:
inputting the eigenvalue matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
obtaining a corresponding entropy value according to the characteristic value of the information of each multimedia resource and the characteristic value of the feedback information of the user;
and determining an evaluation result of each multimedia resource by using the class center value of each cluster class and the entropy value corresponding to each multimedia resource.
21. The method according to claim 11, wherein pushing the multimedia resources according to the evaluation result of each multimedia resource comprises:
and pushing the multimedia resources to the user according to the sequencing result of the evaluation result of each multimedia resource.
22. A server, the server comprising:
the acquisition module is used for acquiring the characteristic value of the information of the multimedia resource;
and the processing module is used for determining an evaluation result of the multimedia resource according to the characteristic value of the information and the characteristic value of the user feedback information corresponding to the multimedia resource.
23. The server according to claim 22, wherein the obtaining module is further configured to obtain a feature value of the user feedback information corresponding to the multimedia resource.
24. The server according to claim 22, wherein the determining the evaluation result of the multimedia resource according to the feature value of the information and the feature value of the user feedback information corresponding to the multimedia resource includes:
obtaining the clustering category of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the feedback information of the user;
Obtaining a class center value of the clustering class;
and determining an evaluation result of the multimedia resource by using the class center value.
25. The server according to claim 22, wherein the determining the evaluation result of the multimedia resource according to the feature value of the information and the feature value of the user feedback information corresponding to the multimedia resource includes:
obtaining the clustering category of the multimedia resource according to the characteristic value of the information of the multimedia resource and the characteristic value of the feedback information of the user;
obtaining a central value of the clustering class;
obtaining the characteristic value of the information of the multimedia resource and the entropy value of the characteristic value of the user feedback information;
and determining an evaluation result of the multimedia resource by using the central value and the entropy value.
26. A server, comprising: a processor, a memory and a program stored on the memory and executable on the processor, which when executed by the processor performs the steps of the method according to any one of claims 1 to 10.
27. A server, the server comprising:
the acquisition module is used for acquiring the characteristic value of the information of at least one multimedia resource in the preset time window;
The processing module is used for determining an evaluation result of each multimedia resource according to the characteristic value of the information of each multimedia resource and the characteristic value of the user feedback information corresponding to each multimedia resource;
and the pushing module is used for pushing the multimedia resources according to the evaluation results of the multimedia resources.
28. The server according to claim 27, wherein the obtaining module is configured to determine at least one multimedia resource of a preset time window from a database;
and acquiring personnel characteristic information and/or video characteristic information of each multimedia resource, and acquiring characteristic values of corresponding information according to the quantized result of the personnel characteristic information and/or the video characteristic information of each multimedia resource.
29. The server according to claim 27, wherein the obtaining module is further configured to obtain objective feedback information and/or subjective feedback information of user feedback corresponding to each multimedia resource,
and obtaining the characteristic value of the user feedback information corresponding to each multimedia resource according to the quantization result of the objective feedback information and/or the subjective feedback information corresponding to each multimedia resource.
30. The server according to claim 27, wherein the processing module is further configured to construct a feature value matrix using feature values of information of at least one multimedia resource and feature values of user feedback information corresponding to each multimedia resource;
The transverse quantity of the characteristic value matrix is used for representing characteristic values of information of the same multimedia resource and corresponding characteristic values of user feedback information; or alternatively, the process may be performed,
the longitudinal quantity of the characteristic value matrix is used for representing the characteristic value of the information of the same multimedia resource and the characteristic value of the corresponding user feedback information.
31. The server according to claim 30, wherein the processing module,
the characteristic value matrix is used for inputting the characteristic value matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
determining an evaluation result of each multimedia resource by utilizing various center values;
wherein each cluster category corresponds to one or more multimedia resources.
32. The server according to claim 30, wherein the processing module,
the characteristic value matrix is used for inputting the characteristic value matrix into a clustering model to obtain a clustering category output by the clustering model;
obtaining class center values of all cluster classes;
obtaining a corresponding entropy value according to the characteristic value of the information of each multimedia resource and the characteristic value of the feedback information of the user;
and determining an evaluation result of each multimedia resource by using the class center value of each cluster class and the entropy value corresponding to each multimedia resource.
33. The server according to claim 27, wherein the pushing module is configured to push the multimedia resources to the user according to the ranking result of the evaluation result of each multimedia resource.
34. A server, comprising: a processor, a memory and a program stored on the memory and executable on the processor, which when executed by the processor performs the steps of the method according to any one of claims 11 to 21.
35. A computer readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, implements the steps of the method according to any of claims 1 to 21.
CN202111341553.XA 2021-11-12 2021-11-12 Multimedia resource evaluation and pushing method and server Pending CN116128324A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111341553.XA CN116128324A (en) 2021-11-12 2021-11-12 Multimedia resource evaluation and pushing method and server

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111341553.XA CN116128324A (en) 2021-11-12 2021-11-12 Multimedia resource evaluation and pushing method and server

Publications (1)

Publication Number Publication Date
CN116128324A true CN116128324A (en) 2023-05-16

Family

ID=86308585

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111341553.XA Pending CN116128324A (en) 2021-11-12 2021-11-12 Multimedia resource evaluation and pushing method and server

Country Status (1)

Country Link
CN (1) CN116128324A (en)

Similar Documents

Publication Publication Date Title
US20220035827A1 (en) Tag selection and recommendation to a user of a content hosting service
US11601703B2 (en) Video recommendation based on video co-occurrence statistics
US9552555B1 (en) Methods, systems, and media for recommending content items based on topics
US9044183B1 (en) Intra-video ratings
US10223438B1 (en) System and method for digital-content-grouping, playlist-creation, and collaborator-recommendation
JP6261547B2 (en) Determination device, determination method, and determination program
US20090089265A1 (en) Information processing apparatus, information processing method, and program
CN109511015B (en) Multimedia resource recommendation method, device, storage medium and equipment
US20220107978A1 (en) Method for recommending video content
CN107894998A (en) Video recommendation method and device
CN104615741B (en) Cold-start project recommendation method and device based on cloud computing
WO2023125460A1 (en) Sorting model training method and sorting method
CN114218482A (en) Information pushing method and device
CN111435371A (en) Video recommendation method and system, computer program product and readable storage medium
CN110958472A (en) Video click rate rating prediction method and device, electronic equipment and storage medium
CN106445922B (en) Method and device for determining title of multimedia resource
CN110913249B (en) Program recommendation method and system
CN116128324A (en) Multimedia resource evaluation and pushing method and server
CN112104910B (en) Video searching method, device and system
JP2012059057A (en) Method, device, and program for calculating viewing probability, and method, device, and program for recommending viewing plan
CN108260007B (en) Program recommendation method and program recommendation system
US20240119468A1 (en) Scoring and recommending a media file
Jiang et al. Prediction of movie playback based on ordinal support vector machine classification
US20240098317A1 (en) Systems and methods for associating program actors with program genres
CN107704477B (en) Multimedia content classification system and method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination