CN110263188B - Media data scoring method, device and equipment - Google Patents
Media data scoring method, device and equipment Download PDFInfo
- Publication number
- CN110263188B CN110263188B CN201910468134.9A CN201910468134A CN110263188B CN 110263188 B CN110263188 B CN 110263188B CN 201910468134 A CN201910468134 A CN 201910468134A CN 110263188 B CN110263188 B CN 110263188B
- Authority
- CN
- China
- Prior art keywords
- media data
- user
- score
- label
- tag
- 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.)
- Active
Links
Images
Classifications
-
- 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
- G06F16/43—Querying
- G06F16/435—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F16/43—Querying
- G06F16/435—Filtering based on additional data, e.g. user or group profiles
- G06F16/437—Administration of user profiles, e.g. generation, initialisation, adaptation, distribution
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a media data scoring method, a device and equipment, wherein the method comprises the following steps: acquiring a first favorite label of a first user in a platform containing a plurality of media data, wherein the first favorite label is a favorite label selected from a plurality of labels by the first user; acquiring an information set associated with the first media data, wherein the information set comprises an evaluation tag of at least one second user on the first media data and a score of the at least one second user on the first media data; and determining a predicted score of the first user for the first media data according to the similarity of the first favorite label and the evaluation label and the target score of the first media data. According to the technical scheme, the targeted media data score can be generated according to the user preference, so that the score can reflect the user preference better.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, and a device for scoring media data.
Background
At present, the scores of the media data seen by each user after logging in the media data platform are the same, and the scores of the media data displayed on the media data platform have no pertinence. The scoring result of the final media data is not accurate enough due to the fact that the diversity of users scoring the media data and the difference of interests and hobbies of the users are large, some users may score higher due to the fact that they like certain types of media data, other users may not like certain types of media data and may score lower, and therefore the simple scoring subjectivity is strong, the accuracy of the final scoring is affected, and each user sees the same scoring of the media data and has no pertinence, and the user who does not watch the media data is not convenient to select whether to watch the media data.
Disclosure of Invention
The embodiment of the application provides a media data scoring method, a device and equipment, and solves the problem that the scoring of media data is not targeted.
In a first aspect, a method for scoring media data is provided, comprising:
acquiring a first favorite label of a first user in a platform containing a plurality of media data, wherein the first favorite label is a favorite label selected by the first user from a plurality of labels, and the plurality of labels comprise labels of each media data in the plurality of media data;
acquiring an information set associated with the first media data, wherein the information set comprises an evaluation label of at least one second user on the first media data and a score of the at least one second user on the first media data, and the first media data is media data which is not scored by the first user;
and determining the predicted score of the first user for the first media data according to the similarity of the first preference label and the evaluation label and combining the target score of the first media data, wherein the target score is determined according to the score of each second user in the information set on the first media data.
In the embodiment of the application, the first favorite tag is a favorite tag selected by the first user, the score of the media data is determined according to the similarity between the favorite tag selected by the first user and the evaluation tag of the second user on the media data, and the score is combined with the favorite of the first user, so that the score has pertinence and can more accurately reflect the favorite of the first user on the media data.
With reference to the first aspect, in a possible implementation manner, before the obtaining the information set associated with the first media data, the method further includes: acquiring first user information of the first user; and screening a plurality of second users of which the user information is matched with the first user information from the plurality of users.
With reference to the first aspect, in a possible implementation manner, before the obtaining the information set associated with the first media data, the method further includes: and screening a plurality of second users with the favorite labels matched with the first favorite labels from the plurality of users.
With reference to the first aspect, in a possible implementation manner, the method further includes: receiving the grade of the first user on the second media data; and when the score of the first user on the second media data is higher than a preset score threshold value and the label of the second media data is not included in the first favorite label, adding the label of the second media data to the first favorite label.
With reference to the first aspect, in a possible implementation manner, the method further includes: acquiring N multimedia data watched by the first user, wherein N is an integer greater than 1; if a target evaluation tag exists in the evaluation tags of each of the N multimedia data and the target evaluation tag is not included in the first favorite tag, adding the target evaluation tag to the first favorite tag.
With reference to the first aspect, in a possible implementation manner, the determining a predicted score of the first user for the first media data according to the similarity between the first preference tag and the rating tag and the target score includes: determining the predictive score if the similarity is greater than or equal to a first similarity threshold, the predictive score being greater than or equal to the target score; determining the prediction score if the similarity is greater than a second similarity threshold and less than the first similarity threshold, the prediction score being less than the target score, the first similarity threshold being greater than the second similarity threshold; determining the prediction score if the similarity is less than the second similarity threshold, the prediction score being the target score.
With reference to the first aspect, in a possible implementation manner, the tags include at least one of attributes of each of the media data and information included in each of the media data.
In a second aspect, there is provided a media data scoring apparatus comprising:
a tag obtaining module, configured to obtain a first favorite tag of a first user in a platform including a plurality of media data, where the first favorite tag is a favorite tag selected by the first user from a plurality of tags, and the plurality of tags include tags of each media data in the plurality of media data;
an information set obtaining module, configured to obtain an information set associated with the first media data, where the information set includes an evaluation tag of the first media data by at least one second user and a score of the first media data by the at least one second user, and the first media data is media data that has not been scored by the first user;
and the score prediction module is used for determining the prediction score of the first user for the first media data according to the similarity of the first favorite label and the evaluation label and combining a target score of the first media data, wherein the target score is determined according to the score of each second user in the information set for the first media data.
In a third aspect, there is provided a media data scoring apparatus, comprising a processor, a memory, and an input/output interface, the processor, the memory, and the input/output interface being connected to each other, wherein the input/output interface is used for inputting or outputting data, the memory is used for storing application program codes for the media data scoring apparatus to execute the method, and the processor is configured to execute the method of the first aspect.
In a fourth aspect, there is provided a computer storage medium storing a computer program comprising program instructions which, when executed by a processor, cause the processor to perform the method of the first aspect described above.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
Fig. 1 is a schematic structural diagram of a media data scoring system according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a media data scoring method according to an embodiment of the present invention;
fig. 3 is a schematic flowchart of another media data scoring method according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a media data scoring apparatus according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of a media data scoring apparatus according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The scheme of the embodiment of the application is suitable for a scene that the user scores the media data, under the condition that the user does not score the media data (hereinafter referred to as an unviewed user), the similarity of a first preference label set by the unviewed user and an evaluation label of the media data evaluated and scored user (hereinafter referred to as a watched user) in an information set on the media data is determined according to the similarity of the evaluation label of the unviewed user and the evaluation label of the unviewed user on the media data, and the prediction score of the unviewed user on the media data is determined according to the evaluation of the watched user on the media data in the information set.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a media data scoring system according to an embodiment of the present disclosure, and as shown in the figure, the media data scoring system includes a first user terminal 101, a media data scoring server 102, and a second user terminal 103. The first user terminal 101 may be configured to perform user interaction, respond to user operation, obtain an evaluation tag and a score of the first user on the media data, and send the evaluation tag and the score to the media data scoring server 102, and so on. The media data scoring server 102 may be configured to receive the rating labels and the scores of the media data sent by the first user terminal 101 and the second user terminal 103; and may also be used to determine a prediction score for the first user for media data that the first user has not scored and send the prediction score to the first user terminal 101, and so on. The second user terminal 103 may be configured to perform user interaction, respond to user operation, obtain an evaluation tag and a score of the first user on the media data, and send the evaluation tag and the score to the media data scoring system 102, and so on. The first user terminal 101 and the second user terminal 103 may be, for example, mobile phones, computers, tablet computers, and the like. The number of the first user terminals 101 may be plural; the number of the second user terminals 103 may also be multiple; the media data scoring server 102 may be one or a group of media data scoring servers, and a plurality of media data scoring servers in the group of media data scoring servers may cooperate to complete processing of an evaluation tag and a score of the media data sent by the user terminal by the user.
Firstly, a first user selects a first favorite tag in a platform containing a plurality of media data through a first user terminal 101, a second user evaluates the first media data through a second user terminal 103 to obtain an evaluation tag and a score, and sends the evaluation tag and the score to a media data scoring server 102; then, the media data scoring system 102 obtains the first favorite tag and the information set associated with the first media data, determines a prediction score of the first user for the first media data according to the similarity between the first favorite tag and the evaluation tags of the second users in the information set associated with the first media data, and combines a target score determined according to the scores of the first media data by the respective second users in the information set, and sends the prediction score to the first user terminal 101, so as to obtain the prediction score of the first user for the first media data.
The media data scoring device according to the embodiment of the present application may be a device with processing capability, for example: devices such as tablet computers, mobile phones, electronic readers, personal Computers (PCs), notebook computers, and servers; or may be a media data scoring module embedded in a media data scoring system. The embodiment of the present application does not limit this.
Referring to fig. 2, fig. 2 is a schematic flowchart of a media data scoring method according to an embodiment of the present application, and as shown in the drawing, the method includes:
s101, acquiring a first favorite label of a first user in a platform containing a plurality of media data, wherein the first favorite label is a favorite label selected by the first user from the plurality of labels, and the plurality of labels comprise labels of each media data in the plurality of media data.
In one possible scenario, the media data may be a video, such as a movie, a television show, a short video, a variety program, etc.; the platform of the media data may be a platform for displaying the media data and the information related to the media data, and may be, for example, a media data player, such as an XX video player, or a plug-in for playing the media data, etc.
The plurality of tags may include attributes of the media data (i.e., characteristics the media data has), such as a place of production of the media data (e.g., domestic, american, korean, etc.), a subject of the media data (e.g., excitement, campus, history, etc.), a type of the media data (e.g., suspense, thriller, comedy, etc.). For example, the plurality of tags may include a homemade, a american drama, a korean drama, a juxtapose, a campus, a history, a suspense, a thriller, a comedy, etc., and the first preference tag may be a homemade, a campus, a comedy, etc.
Optionally, the plurality of tags may also include information contained in the media data, for example, the information contained in the media data may include which actors are in the media data, the type of scenario, whether there is a reversal in the media data, the duration of the media data is greater than or less than a certain duration, and so on.
The first user is a user who does not score the first media data, the first user may pre-select at least one tag from the plurality of tags when initially using the platform of the media data, and the first user may also change the tag selected in the platform of the media data at any time. The first favorite tag of the first user comprises a tag selected by the first user last time, and optionally, the first favorite tag further comprises a tag added according to other modes, for example, when the score of the first user on certain media data is higher than a preset score threshold value and the tag of the second media data is not included in the first favorite tag, the tag of the second media data is added to the first favorite tag. If a target evaluation tag exists in the evaluation tags of each of the N multimedia data viewed by the first user and the target evaluation tag is not included in the first favorite tag, adding the target evaluation tag to the first favorite tag.
S102, an information set associated with the first media data is obtained, the information set comprises an evaluation label of at least one second user to the first media data and a score of the at least one second user to the first media data, and the first media data are media data which are not scored by the first user.
Here, the second user is a user who has rated and scored the first media data, and the rating tag of the second user on the first media data may be obtained through rating of the second user on the first media data. For example, the second user may rate the first media data one or more times to obtain at least one rating tag for the first media data by the second user. As another example, the second user may only have one score for the score of the first media data.
Alternatively, the rating tag of the second user for the first media data may be a tag selected by the second user in a platform of the media data, and the tag may be a preset tag of the first media data. The information set is a set formed by the evaluation labels of the first media data by the second users and the scores of the first media data by the second users, wherein the second users are related to the first media data.
And S103, determining the prediction score of the first user for the first media data according to the similarity of the first preference label and the evaluation label and combining the target score of the first media data, wherein the target score is determined according to the score of each second user in the information set for the first media data.
Here, since each second user has one corresponding rating for the first media data, a plurality of second users correspond to a plurality of corresponding ratings.
In one possible implementation, the target score may be an average of scores of the first media data by respective second users of the plurality of second users. For example, if there are 5 second users scoring the first media data respectively as 5.0, 6.0, 7.0, 8.0, and 9.0, the target score is (5.0 +6.0+7.0+8.0+ 9.0)/5 =7.0.
In another possible implementation, the target score may be an average of scores of the first media data by active users of the plurality of second users. Here, the valid user is a user of the second user who intersects the first favorite tag among the rating tags for the first media data. The evaluation tags of the second user to the first media data include one or more tags in the first favorite tags. For example, the first favorite tag includes a homemade tag, a campus tag, and a comedy tag, and the rating tag of the second user for the first media data includes a campus tag, a rich plot, a good rating, and the like, so that the second user is a valid user. For example, if the scores of the 5 valid users on the first media data are 5.5, 6.0, 6.5, 8.0, and 8.5, respectively, the target score is (5.5 +6.0+6.5+8.0+ 8.5)/5 =6.9.
In another possible implementation manner, the target score may be a weighted average of the score of the common second user, the score weight of the common second user, the score of the valid user, and the score weight of the valid user in the plurality of second users. It is understood that the plurality of second users consists of the normal second user and the active user (i.e., the normal second user except the active user). Here, the valid user is a user of the second user who has an intersection with the first favorite tag in the rating tag of the first media data, and the normal second user is a user other than the valid user (i.e., the normal second user has no intersection with the first favorite tag in the rating tag of the first media data).
For example, the scores of the first media data by 2 common second users in the 5 second users are 5.0, 6.0, respectively, the scores of the first media data by 3 valid users are 7.0, 8.0, 9.0, respectively, the score weight of the common second user is 0.2, and the score weight of the valid user is 0.8, then the target score is (5.0 × 0.2+6.0 × 0.2)/2 + (7.0 × 0.8+8.0 × 0.8+9.0 × 0.8)/3 =1.1+6.4=7.5.
Here, the similarity is a matching degree between the first favorite label and the evaluation label, that is, the evaluation label includes the first favorite label, which indicates that the first favorite label matches the evaluation label.
In an alternative, the similarity is determined based on the number of identical ones of the first preference label and the rating label. For example, if the first favorite label is "domestic product, campus, comedy, drama without dragging or reversing", and the evaluation label is "domestic product, campus, comedy, drama without dragging or reversing", it indicates that the similarity between the first favorite label and the evaluation label is high, and the similarity may be any value from 80% to 100%; if the evaluation label includes at least one of "domestic product, campus, comedy, plot not draggy, no inversion", the similarity may be any one value among 20% to 79%; if the evaluation label does not include any of "domestic product, campus, comedy, drama not draggy, and no inversion", the similarity may be any value from 0 to 19%.
In another alternative, the similarity is determined according to whether the first preference label and the label meaning in the evaluation label are consistent. For example, if the evaluation label is "american, campus, comedy, drama not dragged and not reversed", it indicates that the evaluation label includes a first favorite label, and since the label does not have a meaning with the first favorite label, for example, the first favorite label is "drama not dragged" and the evaluation label is "drama dragged" and indicates that the label does not have a meaning, the similarity between the evaluation label and the first favorite label is low, and may be any value of 20% to 79%, for example, 25%; if the evaluation label is "rich scenario, complex character relationship, and wonderful plot", it indicates that the evaluation label does not include the first preference label, and the similarity may be any value from 0 to 19%, such as 0.
Determining the predicted score of the first user for the first media data according to the similarity between the first preference tag and the rating tag and the target score of the first media data may be performed as follows:
in the first case, in the case where the similarity is greater than or equal to the first similarity threshold, a prediction score is determined, the prediction score being greater than or equal to the target score. In this case, if the first similarity threshold is 80%, the target score is 6.8, and the similarity is 90%, the prediction score may be 8.0 or other value greater than or equal to 6.8.
In the second case, when the similarity is greater than the second similarity threshold and less than the first similarity threshold, the prediction score is determined, the prediction score is less than the target score, and the first similarity threshold is greater than the second similarity threshold. In this case, if the first similarity threshold is 80%, the second similarity threshold is 20%, the target score is 6.8, and the similarity is 50%, the prediction score may be 6.0 or other positive number less than 6.8.
And in the third case, determining a prediction score under the condition that the similarity is smaller than the second similarity threshold, wherein the prediction score is a target score. In this case, if the second similarity threshold is 20%, the target score is 6.8, and the similarity is 0%, the prediction score is the target score, i.e., 6.8.
In one possible case, in a case where the first media data is not scored by the first user, the display score of the first media data in the terminal of the first user is a predicted score, and in a case where the first media data has been scored by the first user, the display score of the first media data in the terminal of the first user is updated to the score of the first media data by the first user.
In the embodiment of the application, the first favorite tag is the favorite tag selected by the first user, the score of the media data is determined according to the similarity between the favorite tag selected by the first user and the evaluation tag of the second user on the media data, and the score is combined with the favorite of the first user, so that the score has pertinence and can more accurately reflect the favorite degree of the first user on the media data.
Referring to fig. 3, fig. 3 is a schematic flow chart of another media data scoring method according to an embodiment of the present invention, as shown in the figure, the method includes:
s201, obtain a first favorite tag of a first user in a platform containing a plurality of media data, where the first favorite tag is a favorite tag selected by the first user from a plurality of tags, and the plurality of tags include tags of each media data in the plurality of media data.
Here, for a method for specifically acquiring the first favorite label of the first user in the platform including multiple media data, reference may be made to the description of step S101 in the embodiment corresponding to fig. 2, which is not described herein again.
S202, first user information of the first user is obtained.
Here, the first user information of the first user may include gender, age, occupation, cultural degree, etc. of the first user. For example, the first user information of the first user may be as shown in table 1:
first user gender | Age (age) | Occupation of the world | Degree of culture |
Female | 25 | Teacher's teacher | This section |
Table 1: first user information table (example) of first user
S203, a plurality of second users of which the user information is matched with the first user information are screened out from the plurality of users.
Here, the pre-saved user information may be acquired from a platform of the media data, and the user information may include a gender, an age, an occupation, a cultural degree, and the like of the second user. For example, the user information may be as shown in table 2:
user' s | Gender of user | Age (age) | Occupation of the world | Degree of culture |
User 1 | Woman | 28 | Teacher's teacher | This section |
User 2 | For male | 25 | Software engineer | This section |
User 3 | Female | 22 | Teacher's teacher | University Special discipline |
Table 2: user information table (example)
Here, the users whose user information in table 2 matches the first user information in table 1 are user 1 and user 3. Here, since the ages of the user 1 and the user 3 are different from the age of the first user, the sexes are the same, and the profession and the academic history are close, it can be considered that the user information of the user 1 and the user information of the user 3 match the first user information. Therefore, the user 1 and the user 3 can be considered to be similar to the first user, and the prediction score of the first user on the first media data can be determined by the evaluation label and the score of the user similar to the preference of the first user and combining the target score to enable the prediction score to reflect the preference of the first user better.
S204, a plurality of second users with the favorite labels matched with the first favorite labels are screened out from the plurality of users.
Here, the first favorites label may be as shown in table 3:
first favorite label | Made in China | Inspiring will | Workplace | Playback of songs |
Table 3: first list of favorites labels (example)
Favorites labels can be as shown in table 4:
table 4: favorite labels table (example)
Here, as can be seen from tables 3 and 4, if the favorite label 3 is the same as the first favorite label, the favorite label 3 matches the first favorite label; favorites label 4 is similar to the first favorites label, and favorites label 4 may be considered to match the first favorites label. It can therefore be assumed that the second user 3 and the second user 4 have similar preferences to the first user, and that determining the prediction score for the first media data by the first user based on the rating labels and scores of the users having similar preferences to the first user, in combination with the target score, may make the prediction score more reflective of the first user's preferences.
Wherein, step S202 and step S203 are the first way of screening a plurality of second users; step S204 is a second manner for screening a plurality of second users, where the first manner and the second manner are two parallel manners, and one of the manners may be selected to be executed in the embodiment shown in fig. 3.
S205, an information set associated with the first media data is obtained, wherein the information set comprises an evaluation label of at least one second user to the first media data and a score of the at least one second user to the first media data, and the first media data is media data which is not scored by the first user.
And S206, determining the prediction score of the first user for the first media data according to the similarity of the first preference label and the evaluation label and combining the target score of the first media data, wherein the target score is determined according to the score of each second user in the information set for the first media data.
Here, the specific implementation manner of steps S205 to S206 may refer to the description of steps S102 to S103 in the embodiment corresponding to fig. 2, and is not described herein again.
And S207, receiving the scores of the first user on the second media data.
Here, the second media data is the media data that is evaluated and scored by the first user, the number of the second media data may be multiple, and each second media data corresponds to a score of the first user.
S208, when the score of the first user on the second media data is higher than the preset score threshold and the tag of the second media data is not included in the first favorite tag, adding the tag of the second media data to the first favorite tag.
Wherein the label of the second media data includes the following two cases:
in the first case, the tag of the second media data is a tag of the second media data preset in the platform of the media data. Such as attributes of the media data.
In the second case, the tag of the second media data is an evaluation tag obtained by evaluating the second media data by the third user.
Here, the third user may include a user determined in three ways:
in the first mode, the third users are all users who have rated the second media data.
In a second manner, the third user is a user who has rated the second media data among the multiple second users whose user information matches the first user information, where the multiple second users are determined in step S203.
In a third manner, the third user is a user who has rated the second media data among the plurality of second users whose favorite labels are matched with the first favorite labels, among the plurality of users determined in step S204.
Here, the first favorite tag includes a tag selected by the first user and a tag that the first user may like and is obtained by the platform of the media data. The label selected by the first user is based on the label newly selected by the first user; the tags that the first user may like acquired by the platform of the media data may be obtained through step S208.
For example, in a case that the second media data evaluated and scored by the first user does not include the first favorite tag, the second media data with a higher score of the first user in the platform of the media data may be acquired, and the tag of the second media data may be added to the tag that the first user may like acquired by the platform of the media data in the first favorite tag. The tag of the second media data is not included in the first favorites tag i.e. the second media data does not have the features in the first favorites tag. For example, if the first preference label is "comedy, rich drama, brainstorming, and reverse", and the second media data is "science fiction, suspicion, and drama", it is considered that the label of the second media data is not included in the first preference label.
Here, the preset score threshold may be any positive number such as 7.0, 7.5, 8.0, and the like. For example, if 3 pieces of second media data are respectively the second media data 1, the second media data 2, and the second media data 3, and none of the tags of the 3 pieces of second media data is included in the first favorite tag, the preset score threshold is 6.5, and the scores of the first user for the second media data 1, 2, and 3 are respectively 5.0, 7.0, and 8.0, the tags of the second media data 2 and the second media data 3 are added to the tags that the first user may like and are obtained by the platform of the media data in the first favorite tag.
Optionally, after step S201 is executed, N multimedia data viewed by the first user may also be obtained, where N is an integer greater than 1; if a target evaluation tag exists in the evaluation tags of each of the N multimedia data and the target evaluation tag is not included in the first favorite tag, adding the target evaluation tag to the first favorite tag. For example, 10 consecutive movies watched by the first user are captured, and if the rating labels all contain a foreign label, this foreign label is added to the first favorites label.
In the embodiment of the application, a second user with user information matched with the first user information can be screened from a plurality of users; or screening a second user with a favorite label matched with the first favorite label from the favorite labels, and determining a prediction score according to the similarity between the evaluation label of the second user and the first favorite label and the target score, wherein the prediction score determined according to the evaluation label of the second user and the score can be more accurate and can reflect the favorite of the first user better because the second user and the first user have the same characteristics; the method comprises the steps of obtaining the score of a first user on media data under the condition that the first user is detected to evaluate and the scored media data do not contain a first favorite tag, and adding the tag of the media data into the first favorite tag under the condition that the score of the first user on the media data is high, so that the first favorite tag can be updated in real time, and the predicted score obtained according to the similarity of the first favorite tag and the evaluation tag and the target score is more accurate.
The method of the embodiment of the invention is described above, and the apparatus of the embodiment of the invention is described below.
Referring to fig. 4, fig. 4 is a schematic structural diagram of a media data scoring apparatus according to an embodiment of the present application, where the apparatus 40 includes:
a tag obtaining module 401, configured to obtain a first favorite tag of a first user in a platform containing a plurality of media data, where the first favorite tag is a favorite tag selected by the first user from a plurality of tags, and the plurality of tags include tags of each media data in the plurality of media data;
an information set obtaining module 402, configured to obtain an information set associated with the first media data, where the information set includes an evaluation tag of the first media data by at least one second user and a score of the first media data by the at least one second user, and the first media data is media data that has not been scored by the first user;
a score prediction module 403, configured to determine, according to the similarity between the first preference tag and the evaluation tag, and in combination with a target score of the first media data, a prediction score of the first user for the first media data, where the target score is determined according to a score of each second user in the information set for the first media data.
In one possible design, the apparatus 40 further includes:
a user information matching module 404, configured to obtain first user information of the first user;
the user information matching module 404 is further configured to screen a plurality of second users from the plurality of users, where the user information matches the first user information.
In one possible design, the apparatus 40 further includes:
a favorite label matching module 405, configured to screen out a plurality of second users whose favorite labels match the first favorite labels from the plurality of users.
In one possible design, the apparatus 40 further includes:
a tag adding module 406, configured to receive a rating of the second media data by the first user; adding a label of second media data to the first favorite label when the score of the first user on the second media data is higher than a preset score threshold and the label of the second media data is not included in the first favorite label.
In a possible design, the tag adding module 406 is further configured to obtain N multimedia data that the first user has viewed, where N is an integer greater than 1; if a target evaluation tag exists in the evaluation tags of each of the N multimedia data and the target evaluation tag is not included in the first favorite tag, adding the target evaluation tag to the first favorite tag.
In one possible design, the score prediction module 403 is specifically configured to:
determining the prediction score if the similarity is greater than or equal to a first similarity threshold, the prediction score being greater than or equal to the target score;
determining the prediction score if the similarity is greater than a second similarity threshold and less than the first similarity threshold, the prediction score being less than the target score, the first similarity threshold being greater than the second similarity threshold;
determining the prediction score if the similarity is less than the second similarity threshold, the prediction score being the target score.
In one possible design, the plurality of tags includes at least one of an attribute of each of the plurality of media data and information contained in each of the plurality of media data.
It should be noted that, for the content that is not mentioned in the embodiment corresponding to fig. 4, reference may be made to the description of the method embodiment, and details are not described here again.
In the embodiment of the application, because the first favorite label is the favorite label selected by the first user, the score of the media data is determined according to the similarity between the favorite label selected by the first user and the evaluation label of the second user on the media data, and the score is combined with the favorite of the first user, the score has pertinence and can more accurately reflect the favorite degree of the first user on the media data; the second user with the user information matched with the first user information can be screened out from the plurality of users; or screening a second user with a favorite label matched with the first favorite label from the favorite labels, and determining a prediction score according to the similarity between the evaluation label of the second user and the first favorite label and the target score, wherein the prediction score determined according to the evaluation label of the second user and the score can be more accurate and can reflect the favorite of the first user better because the second user and the first user have the same characteristics; the method comprises the steps of obtaining the score of a first user on media data under the condition that the first user is detected to evaluate and the scored media data do not contain a first favorite tag, and adding the tag of the media data into the first favorite tag under the condition that the score of the first user on the media data is high, so that the first favorite tag can be updated in real time, and the predicted score obtained according to the similarity of the first favorite tag and the evaluation tag and the target score is more accurate.
Referring to fig. 5, fig. 5 is a schematic diagram illustrating a component structure of a media data scoring apparatus according to an embodiment of the present application, where the apparatus 50 includes a processor 501, a memory 502, and an input/output interface 503. The processor 501 is connected to the memory 502 and the input/output interface 503, for example, the processor 501 may be connected to the memory 502 and the input/output interface 503 through a bus.
The processor 501 is configured to support the media data scoring apparatus to perform corresponding functions in the media data scoring method described in fig. 2-3. The processor 501 may be a Central Processing Unit (CPU), a Network Processor (NP), a hardware chip, or any combination thereof. The hardware chip may be an Application Specific Integrated Circuit (ASIC), a Programmable Logic Device (PLD), or a combination thereof. The PLD may be a Complex Programmable Logic Device (CPLD), a field-programmable gate array (FPGA), a General Array Logic (GAL), or any combination thereof.
The memory 502 is used to store program codes and the like. Memory 502 may include Volatile Memory (VM), such as Random Access Memory (RAM); the memory 502 may also include a non-volatile memory (NVM), such as a read-only memory (ROM), a flash memory (flash memory), a hard disk (HDD) or a solid-state drive (SSD); the memory 502 may also comprise a combination of the above kinds of memories.
The input/output interface 503 is used for inputting or outputting data.
The processor 501 may call the program code to perform the following operations:
acquiring a first favorite label of a first user in a platform containing a plurality of media data, wherein the first favorite label is a favorite label selected by the first user from a plurality of labels, and the plurality of labels comprise labels of each media data in the plurality of media data;
acquiring an information set associated with the first media data, wherein the information set comprises an evaluation label of at least one second user on the first media data and a score of the at least one second user on the first media data, and the first media data is media data which is not scored by the first user;
and determining the predicted score of the first user for the first media data according to the similarity of the first preference label and the evaluation label and combining the target score of the first media data, wherein the target score is determined according to the score of each second user in the information set on the first media data.
It should be noted that, implementation of each operation may also correspond to the corresponding description with reference to the foregoing method embodiment; the processor 501 may also cooperate with the input-output interface 503 to perform other operations in the above-described method embodiments.
Embodiments of the present application also provide a computer storage medium storing a computer program comprising program instructions which, when executed by a computer, cause the computer to perform the method according to the aforementioned embodiments, the computer may be part of the aforementioned media data scoring device. Such as processor 501 described above.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by a computer program, which may be stored in a computer readable storage medium and executed to implement the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a ROM, a RAM, or the like.
The above disclosure is only for the purpose of illustrating the preferred embodiments of the present application and is not to be construed as limiting the scope of the present application, so that the present application is not limited thereto, and all equivalent variations and modifications can be made to the present application.
Claims (9)
1. A method for scoring media data, comprising:
acquiring a first favorite label of a first user in a platform containing a plurality of media data, wherein the first favorite label is a favorite label selected by the first user from a plurality of labels, and the plurality of labels comprise labels of each media data in the plurality of media data;
acquiring an information set associated with first media data, wherein the information set comprises an evaluation label of at least one second user on the first media data and a score of the at least one second user on the first media data, and the first media data is media data which is not scored by the first user;
determining a predicted score of the first user for the first media data according to the similarity of the first favorite label and the evaluation label and combining a target score of the first media data, wherein the target score is determined according to the score of each second user in the information set on the first media data;
the determining a predicted score of the first user for the first media data according to the similarity of the first preference label and the rating label and in combination with the target score comprises: determining the prediction score if the similarity is greater than or equal to a first similarity threshold, the prediction score being greater than or equal to the target score; determining the prediction score if the similarity is greater than a second similarity threshold and less than the first similarity threshold, the prediction score being less than the target score, the first similarity threshold being greater than the second similarity threshold; determining the prediction score if the similarity is less than the second similarity threshold, the prediction score being the target score.
2. The method of claim 1, wherein before the obtaining the set of information associated with the first media data, further comprising:
acquiring first user information of the first user;
and screening a plurality of second users of which the user information is matched with the first user information from the plurality of users.
3. The method of claim 1, wherein before the obtaining the set of information associated with the first media data, further comprising:
and screening a plurality of second users with the preference labels matched with the first preference label from the plurality of users.
4. The method of claim 1, further comprising:
receiving the grade of the first user on the second media data;
adding a label of second media data to the first favorite label when the score of the first user on the second media data is higher than a preset score threshold and the label of the second media data is not included in the first favorite label.
5. The method of claim 1, further comprising:
acquiring N multimedia data watched by the first user, wherein N is an integer greater than 1;
if a target evaluation tag exists in the evaluation tags of each of the N multimedia data and the target evaluation tag is not included in the first favorite tag, adding the target evaluation tag to the first favorite tag.
6. The method of any one of claims 1-5, wherein the plurality of tags includes at least one of attributes of each of the plurality of media data and information contained in each of the plurality of media data.
7. A media data scoring apparatus, comprising:
a tag obtaining module, configured to obtain a first favorite tag of a first user in a platform including a plurality of media data, where the first favorite tag is a favorite tag selected by the first user from a plurality of tags, and the plurality of tags include tags of each media data in the plurality of media data;
the information set acquisition module is used for acquiring an information set related to first media data, wherein the information set comprises an evaluation label of at least one second user on the first media data and a score of the at least one second user on the first media data, and the first media data is media data which is not scored by the first user;
a score prediction module, configured to determine, according to a similarity between the first preference tag and the evaluation tag, and in combination with a target score of the first media data, a prediction score of the first user for the first media data, where the target score is determined according to a score of each second user in the information set for the first media data;
the determining a predicted score of the first user for the first media data according to the similarity of the first preference label and the rating label and in combination with the target score comprises: determining the prediction score if the similarity is greater than or equal to a first similarity threshold, the prediction score being greater than or equal to the target score; determining the prediction score if the similarity is greater than a second similarity threshold and less than the first similarity threshold, the prediction score being less than the target score, the first similarity threshold being greater than the second similarity threshold; determining the prediction score if the similarity is less than the second similarity threshold, the prediction score being the target score.
8. A media data rating device comprising a processor, a memory and an input-output interface, the processor, the memory and the input-output interface being interconnected, wherein the input-output interface is used for inputting or outputting data, the memory is used for storing program code, and the processor is used for calling the program code to execute the method according to any one of claims 1 to 6.
9. A computer storage medium, characterized in that it stores a computer program comprising program instructions which, when executed by a processor, cause the processor to carry out the method according to any one of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910468134.9A CN110263188B (en) | 2019-05-29 | 2019-05-29 | Media data scoring method, device and equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910468134.9A CN110263188B (en) | 2019-05-29 | 2019-05-29 | Media data scoring method, device and equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110263188A CN110263188A (en) | 2019-09-20 |
CN110263188B true CN110263188B (en) | 2023-03-28 |
Family
ID=67916219
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910468134.9A Active CN110263188B (en) | 2019-05-29 | 2019-05-29 | Media data scoring method, device and equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110263188B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105630880A (en) * | 2015-12-17 | 2016-06-01 | 东软集团股份有限公司 | Score data prediction method and apparatus |
CN108737859A (en) * | 2018-05-07 | 2018-11-02 | 华东师范大学 | Video recommendation method based on barrage and device |
CN109388760A (en) * | 2017-08-03 | 2019-02-26 | 腾讯科技(北京)有限公司 | Recommend label acquisition method, media content recommendations method, apparatus and storage medium |
CN109461016A (en) * | 2018-09-10 | 2019-03-12 | 平安科技(深圳)有限公司 | Data methods of marking, device, computer equipment and storage medium |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050246740A1 (en) * | 2004-05-03 | 2005-11-03 | Teraci Richard D | Apparatus and method for evaluating media |
US10531139B2 (en) * | 2016-11-18 | 2020-01-07 | Byung-In Kim | System and method for recommending media |
-
2019
- 2019-05-29 CN CN201910468134.9A patent/CN110263188B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105630880A (en) * | 2015-12-17 | 2016-06-01 | 东软集团股份有限公司 | Score data prediction method and apparatus |
CN109388760A (en) * | 2017-08-03 | 2019-02-26 | 腾讯科技(北京)有限公司 | Recommend label acquisition method, media content recommendations method, apparatus and storage medium |
CN108737859A (en) * | 2018-05-07 | 2018-11-02 | 华东师范大学 | Video recommendation method based on barrage and device |
CN109461016A (en) * | 2018-09-10 | 2019-03-12 | 平安科技(深圳)有限公司 | Data methods of marking, device, computer equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN110263188A (en) | 2019-09-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11941912B2 (en) | Image scoring and identification based on facial feature descriptors | |
CN109819284B (en) | Short video recommendation method and device, computer equipment and storage medium | |
US10671670B2 (en) | Graph based content browsing and discovery | |
CN106326391B (en) | Multimedia resource recommendation method and device | |
CN105574045B (en) | Video recommendation method and server | |
US20200192905A1 (en) | Methods, systems, and media for presenting related media content items | |
CN110704677B (en) | Program recommendation method and device, readable storage medium and terminal equipment | |
US20170169062A1 (en) | Method and electronic device for recommending video | |
CN112818224A (en) | Information recommendation method and device, electronic equipment and readable storage medium | |
CN111681042A (en) | Advertisement recommendation method, server and terminal equipment | |
CN104967690A (en) | Information push method and device | |
CN115455280A (en) | Recommendation list determining method and server | |
CN111046304B (en) | Data searching method and device | |
AU2020104435A4 (en) | Method and apparatus for video recommendation, and refrigerator with screen | |
US20140289334A1 (en) | System and method for recommending multimedia information | |
CN110263188B (en) | Media data scoring method, device and equipment | |
KR101678779B1 (en) | Method for recommending contents using metadata and apparatus for performing the method | |
CN116663505A (en) | Comment area management method and system based on Internet | |
US20170161322A1 (en) | Method and electronic device for searching resource | |
CN115795158A (en) | Material sorting and neural network training method, device, equipment and storage medium | |
CN112131502A (en) | Data processing method, data processing apparatus, electronic device, and medium | |
KR102323424B1 (en) | Rating Prediction Method for Recommendation Algorithm Based on Observed Ratings and Similarity Graphs | |
CN113360681B (en) | Method, device, electronic equipment and storage medium for determining recommendation information | |
CN105379297A (en) | Interfacing a television with a second device | |
CN113127750B (en) | Information list generation method and device, storage medium and electronic equipment |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |