CN111324755A - Label determining method and device, electronic equipment and storage medium - Google Patents
Label determining method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111324755A CN111324755A CN202010112968.9A CN202010112968A CN111324755A CN 111324755 A CN111324755 A CN 111324755A CN 202010112968 A CN202010112968 A CN 202010112968A CN 111324755 A CN111324755 A CN 111324755A
- Authority
- CN
- China
- Prior art keywords
- work
- label
- work set
- user
- target
- 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
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
- G06F16/45—Clustering; Classification
-
- 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
- G06F16/48—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/483—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Probability & Statistics with Applications (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The disclosure relates to a label determination method, a label determination device, an electronic device and a storage medium. The label determination method comprises the following steps: acquiring a first work set corresponding to a user account, and extracting average characteristics of first works in the first work set; the first work in the first work set is a representative work corresponding to the user account; obtaining a target work and extracting the characteristics of the target work; calculating a feature difference value between the features of the target work and the average features; the method comprises the steps of obtaining a user portrait label corresponding to a user account, and determining the user portrait label as a label of a target work with a characteristic difference value smaller than a preset threshold value. According to the embodiment of the disclosure, when the characteristic difference value is smaller than the preset threshold value, the user portrait label corresponding to the user account is determined as the label of the corresponding target work, so that the phenomenon that the target work and the label are not applicable due to the fact that the user portrait label is directly determined as the label of the target work can be avoided, and the consistency of the content of the target work and the corresponding label is ensured.
Description
Technical Field
The present disclosure relates to the field of internet technologies, and in particular, to a tag determination method and apparatus, an electronic device, and a storage medium.
Background
With the development of internet technology and the great demand of people on cultural art, people have unprecedented development on sharing life enthusiasm by creating videos, and the watching of short videos and various image works gradually becomes one of important ways for people to receive information in daily life. The label determination of the works uploaded by the user is an important basis for user and work recommendation and advertising. The traditional work label determining method generally comprises the steps of simply screening uploaded works and then directly determining a user portrait as a label of the work uploaded by the user.
However, the content of the work uploaded by the same user is not completely consistent, the user portrait is directly determined as the label of the user work after simple screening, and the label which is not suitable for the work is often given to the corresponding work, so that the label of part of the work is seriously inconsistent with the content.
Disclosure of Invention
The disclosure provides a label determination method, a label determination device, an electronic device and a storage medium, which are used for at least solving the problem that the label of a work is seriously inconsistent with the content in the related technology. The technical scheme of the disclosure is as follows:
according to a first aspect of the embodiments of the present disclosure, there is provided a tag determination method, including:
acquiring a first work set corresponding to a user account, and extracting average characteristics of first works in the first work set; a first work in the first work set is a representative work corresponding to the user account;
obtaining a target work and extracting the characteristics of the target work;
calculating a feature difference between the features of the target work and the average features;
and acquiring a user portrait label corresponding to the user account, and determining the user portrait label as a label of the target work of which the characteristic difference value is smaller than a preset threshold value.
In an exemplary embodiment, the obtaining the first set of works corresponding to the user account previously includes:
acquiring the first work set, and labeling each first work in the first work set to obtain first labeling information;
and determining the first labeling information as a user portrait label corresponding to the user account.
In an exemplary embodiment, the obtaining the first set of works corresponding to the user account includes:
acquiring a second work set corresponding to the user account; a second work in the second work set is a work corresponding to the user account;
screening the second work set according to the playing amount corresponding to each second work in the second work set to obtain a third work set;
extracting the characteristics of each third work in the third work set;
and clustering the characteristics of the third works, and screening the third work set according to clustering results to obtain the first work set.
In an exemplary embodiment, the screening the second work set according to a playing amount corresponding to each second work in the second work set to obtain a third work set includes:
calculating the average playing amount of the second work set according to the playing amount corresponding to each second work in the second work set;
and determining a set formed by the second works of which the playing amount is larger than the average playing amount as the third work set.
In an exemplary embodiment, the clustering the features of the third work and screening the third work set according to the clustering result to obtain the first work set includes:
and clustering the characteristics of the third works, and determining a set formed by the third works with the highest density as the first work set.
In an exemplary embodiment, the extracting the average feature of the first work in the first work set comprises:
obtaining characteristics of each first work in the first work set;
calculating an average characteristic of the first work set according to the characteristic of each first work in the first work set.
In an exemplary embodiment, the obtaining a user portrait label corresponding to the user account, and determining the user portrait label as a label of a target work of which the characteristic difference value is smaller than a preset threshold value, then includes:
marking the target work with the characteristic difference value larger than or equal to the preset threshold value to obtain second marking information;
and determining the second labeling information as the label of the target work with the characteristic difference value larger than or equal to the preset threshold value.
According to a second aspect of the embodiments of the present disclosure, there is provided a tag determination apparatus including:
the system comprises a user work set acquisition unit, a first work set acquisition unit and a second work set acquisition unit, wherein the user work set acquisition unit is configured to acquire a first work set corresponding to a user account and extract average characteristics of first works in the first work set; a first work in the first work set is a representative work corresponding to the user account;
a target work acquisition unit configured to perform acquisition of a target work and extract a feature of the target work;
a feature difference determination unit configured to perform calculating a feature difference between the feature of the target work and the average feature;
and the label determining unit is configured to execute the steps of obtaining a user portrait label corresponding to the user account, and determining the user portrait label as a label of the target work of which the characteristic difference value is smaller than a preset threshold value.
In an exemplary embodiment, the tag determination apparatus further includes a first labeling unit configured to perform:
acquiring the first work set, and labeling each first work in the first work set to obtain first labeling information;
and determining the first labeling information as a user portrait label corresponding to the user account.
In an exemplary embodiment, the user work set obtaining unit is further configured to perform:
acquiring a second work set corresponding to the user account; a second work in the second work set is a work corresponding to the user account;
screening the second work set according to the playing amount corresponding to each second work in the second work set to obtain a third work set;
extracting the characteristics of each third work in the third work set;
and clustering the characteristics of the third works, and screening the third work set according to clustering results to obtain the first work set.
In an exemplary embodiment, the user work set obtaining unit is further configured to perform:
calculating the average playing amount of the second work set according to the playing amount corresponding to each second work in the second work set;
and determining a set formed by the second works of which the playing amount is larger than the average playing amount as the third work set.
In an exemplary embodiment, the user work set obtaining unit is further configured to perform:
and clustering the characteristics of the third works, and determining a set formed by the third works with the highest density as the first work set.
In an exemplary embodiment, the user work set obtaining unit is further configured to perform:
obtaining characteristics of each first work in the first work set;
calculating an average characteristic of the first work set according to the characteristic of each first work in the first work set.
In an exemplary embodiment, the tag determination apparatus further includes a second labeling unit configured to perform:
marking the target work with the characteristic difference value larger than or equal to the preset threshold value to obtain second marking information;
and determining the second labeling information as the label of the target work with the characteristic difference value larger than or equal to the preset threshold value. According to a third aspect of the embodiments of the present disclosure, there is provided an electronic apparatus including:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the tag determination method of the first aspect.
According to a fourth aspect of embodiments of the present disclosure, there is provided a storage medium, wherein instructions that, when executed by a processor of the electronic device, enable the electronic device to perform the tag determination method of the first aspect.
The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects:
the average feature of the first works in the first work set is extracted by obtaining the first work set corresponding to the user account, the average feature is compared with the feature of the target work, a feature difference value between the feature of the target work and the average feature is obtained, the smaller the feature difference value is, the more similar the target work and the first works in the first work set is considered to be, when the feature difference value is smaller than a preset threshold value, a user portrait label corresponding to the user account is obtained, and the user portrait label is determined to be the label of the target work of which the feature difference value is smaller than the preset threshold value. Therefore, the method for determining the label of the target work according to the relation between the specific target work characteristics and the average characteristics of the first work in the first work set can avoid the phenomenon that the target work and the label are not applicable due to the fact that the user portrait is directly determined as the label of the target work, and the consistency of the content of the target work and the corresponding label is guaranteed. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.
Fig. 1 is a flow chart illustrating a method of tag determination according to an example embodiment.
Fig. 2 is a flowchart illustrating one possible implementation of step S100 according to an exemplary embodiment.
Fig. 3 is a flowchart illustrating one possible implementation of step S120 according to an exemplary embodiment.
Fig. 4 is a flowchart illustrating one implementable approach after step S400, according to an example embodiment.
Fig. 5 is a block diagram illustrating a tag determination apparatus according to an example embodiment.
FIG. 6 is a block diagram illustrating an electronic device for tag determination in accordance with an exemplary embodiment.
FIG. 7 is a block diagram illustrating an apparatus for tag determination in accordance with an example embodiment.
Detailed Description
In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
Fig. 1 is a flow chart illustrating a tag determination method according to an exemplary embodiment, as shown in fig. 1, including the steps of:
in step S100, a first work set corresponding to the user account is obtained, and an average feature of the first work in the first work set is extracted; and the first work in the first work set is a representative work corresponding to the user account.
In step S200, a target work is acquired, and features of the target work are extracted.
In step S300, a feature difference between the features of the target work and the average features is calculated.
In step S400, a user portrait label corresponding to the user account is obtained, and the user portrait label is determined as a label of the target work with a characteristic difference value smaller than a preset threshold value.
For example, most of the works uploaded by the user are related to gourmet, but at the same time, works with people or tourist scenery as main content may also be uploaded, and related works with people or tourist scenery as main content are not the representative works of the user. The feature refers to a combination of characteristics of the content of a specific work, and can be extracted through a specific model or a network. The target work generally refers to a work to be determined by using a tag uploaded by a corresponding user, and of course, the target work may also be a work uploaded by other users, which is not specifically limited herein. The user portrait is an effective tool for sketching a target user and associating user appeal with a design direction, the user mutual label is marking information corresponding to the user portrait, for example, a user uploading short videos is taken as an example, the user portrait label of the user is generally ' food ', the user portrait label of the user is generally ' tourist landscape ', and when the user uploads the work is generally ' tourist travel-related work, the user portrait label of the user is generally ' tourist landscape ', and the description about ' food ' and ' tourist landscape ' is only an exemplary description and is not used for limiting specific labels.
Specifically, a certain number of representative works of a specific user are obtained from works uploaded by the specific user, a set formed by the certain number of representative works is used as a first work set, and the certain number of representative works are input into a preset feature extraction model or a feature extraction network (or other feature extraction modes are available, and the specific feature extraction is not limited at that time). And carrying out weighted average on the extracted features to obtain the average features of the first works in the first work set. After the target work is obtained and the features of the target work are extracted, the difference is made between the features of the target work and the average features of the first works in the first work set, a feature difference value between the features of the target work and the average features of the first works in the first work set is obtained, when the feature difference value is smaller, the fact that the features of the target work are similar to the features of the first works in the first work set is indicated, a user portrait label corresponding to the user account is obtained, and the corresponding user portrait label is determined to be the label of the target work. For example, the first works in the first work set are gourmet works, the more similar the characteristics of the target work are to the first works in the first work set, the higher the possibility that the target work is gourmet works is, and at this time, the user portrait label (gourmet) corresponding to the user account is determined as the label of the target work.
According to the label determining method, the average feature of the first works in the first work set is extracted by obtaining the first work set corresponding to the user account, the average feature is compared with the feature of the target work to obtain the feature difference value between the feature of the target work and the average feature, the smaller the feature difference value is, the more similar the target work is to the first work in the first work set, when the feature difference value is smaller than the preset threshold value, the user portrait label corresponding to the user account is obtained, and the user portrait label is determined to be the label of the target work of which the feature difference value is smaller than the preset threshold value. Therefore, the method for determining the label of the target work according to the relation between the specific target work characteristics and the average characteristics of the first work in the first work set can avoid the phenomenon that the target work and the label are not applicable due to the fact that the user portrait is directly determined as the label of the target work, and the consistency of the content of the target work and the corresponding label is guaranteed.
In an exemplary embodiment, the step S100 is an implementable manner, wherein obtaining the first work set corresponding to the user account includes:
acquiring a first work set, and labeling each first work in the first work set to obtain first labeling information; and determining the first marking information as a user portrait label corresponding to the user account.
Specifically, the first work in the first work set is a representative work corresponding to the user, and after the representative work is subjected to first labeling, the user portrait label of the user account can be determined according to the obtained first labeling information. For example, if a particular user uploads 100 works, 90 of which are related to food, the 90 works can be marked as food, and the user portrait of the user is determined as food, and the user portrait label is also marked as food.
In the above exemplary embodiment, the first label information is obtained by obtaining the first work set and labeling each first work in the first work set, the first label information is determined as the user portrait label corresponding to the user account, a data basis is provided for subsequent label determination of the target work, and finally the label of the target work close to the average feature of the first work in the first work set is determined as the user portrait label.
Fig. 2 is a flowchart illustrating an implementable manner of step S100 according to an exemplary embodiment, where as shown in fig. 2, acquiring a first work set corresponding to a user account includes the following steps:
in step S110, a second work set corresponding to the user account is obtained; and the second works in the second work set are works corresponding to the user account.
In step S120, the second work set is screened according to the playing amount corresponding to each second work in the second work set, so as to obtain a third work set.
In step S130, the feature of each third work in the third work set is extracted.
In step S140, the features of the third work are clustered, and the third work set is screened according to the clustering result, so as to obtain the first work set.
Specifically, according to a corresponding user account, taking a set formed by all works uploaded by a user as a second work set, obtaining a playing amount corresponding to each second work in the second work set, screening the second works in the second work set according to the playing amount, taking the set formed by the screened second works as a third work set, inputting the third works in the third work set into a feature extraction model or a feature extraction network, performing feature extraction to obtain features of each third work in the third work set, and clustering the features of the third works.
Optionally, the features of the third works are clustered, and the set formed by the third works with the highest density is determined as the first work set.
Specifically, the third work set is screened according to the clustering result, and a set formed by one or more groups of third works with the highest clustering density is used as the first work set. Optionally, a specific clustering method is not limited, and for example, DBScan, K-means, and the like may be used to cluster the features of the third work.
In the above exemplary embodiment, a second work set corresponding to the user account is obtained, where the second work in the second work set is a work corresponding to the user account, the second work set is screened according to the playing amount corresponding to each second work in the second work set to obtain a third work set, characteristics of each third work in the third work set are extracted, the characteristics of the third work are clustered, the third work set is screened according to the clustering result, in the process of obtaining the first work set, the first work set is firstly screened according to the playing amount corresponding to the second work, and the characteristics of the screened works are clustered, where the greater the clustering density is, the greater the similarity of the corresponding works is, the set composed of one or more groups of works with the greatest similarity is used as the first work set, so as to reflect characteristics of the works uploaded by the user to the greatest extent, the phenomenon that the target works and the labels are not applicable due to the fact that the user portrait is directly determined as the labels of the target works can be avoided, and the consistency of the content of the target works and the corresponding labels is guaranteed.
Fig. 3 is a flowchart illustrating an implementable manner of step S120 according to an exemplary embodiment, as shown in fig. 3, wherein the step of screening the second work set according to a corresponding playing amount of each second work in the second work set to obtain a third work set includes the following steps:
in step S121, an average playing amount of the second work set is calculated according to the playing amount corresponding to each second work in the second work set.
In step S122, a set of second works whose playback volume is larger than the average playback volume is determined as a third set of works.
Specifically, after the second work set is obtained, the average playing amount of the second work set is calculated according to the playing amount corresponding to each second work in the second work set. Generally, the larger the playing amount of the corresponding works is, the greater the attraction of the corresponding works is, the richer the content is, the types of the works which can represent the corresponding users are better, and the value of recommending and advertising the works is also higher. Therefore, screening the works with larger playing amount is more beneficial to subsequent label determination according to the characteristics of the target works, and on the basis, determining the set formed by the second works with the playing amount corresponding to each second work larger than the average playing amount as a third work set, so as to provide a data basis for subsequently obtaining the first work set with user representativeness.
In the above exemplary embodiment, the average playing amount of the second work set is calculated according to the playing amount corresponding to each second work in the second work set, and the set formed by the second works of which the playing amount is greater than the average playing amount is determined as the third work set, so as to provide a data base for subsequently obtaining the first work set with user representativeness, and finally achieve the purpose of determining the target work label according to the relationship between the specific target work characteristics and the average characteristics of the first works in the first work set.
In an exemplary embodiment, the step S100 is an implementable manner, wherein the extracting the average feature of the first work in the first work set includes:
obtaining the characteristics of each first work in the first work set; an average feature of the set of first works is calculated based on the features of each first work in the set of first works.
Specifically, each first work in the first work set is input into a preset feature extraction model or a feature extraction network (or in other feature extraction manners, the specific time of feature extraction is not limited here), feature extraction is performed, and the extracted features are weighted and averaged to obtain an average feature of the first work in the first work set.
The above-described exemplary embodiment, by obtaining the characteristics of each first work in the first work set; according to the characteristics of each first work in the first work set, the average characteristics of the first work set are calculated, the obtained average characteristics can reflect the characteristics of the corresponding user more comprehensively, the phenomenon that the target work and the label are not applicable due to the fact that the user portrait is directly determined as the label of the target work is avoided, and the consistency of the content of the target work and the corresponding label is guaranteed.
Fig. 4 is a flowchart illustrating an implementable manner after step S400 according to an exemplary embodiment, as shown in fig. 4, where a user portrait label corresponding to a user account is obtained, and the user portrait label is determined as a label of a target work with a characteristic difference smaller than a preset threshold, and then the following steps are included:
in step S410, labeling the target work whose feature difference is greater than or equal to the preset threshold to obtain second labeling information.
In step S420, the second annotation information is determined as the label of the target work whose feature difference is greater than or equal to the preset threshold.
Specifically, when the characteristic difference is greater than or equal to the preset threshold, it is described that the smaller the similarity between the target work and the first work in the first work set is, the larger the deviation between the target work corresponding to the characteristic difference greater than or equal to the preset threshold and the user portrait label of the user is, and the target work cannot be labeled by the user portrait label of the user. At this time, the target work is labeled to obtain second labeling information, and the second labeling information is determined as the label of the target work. For example, the user portrait label of the user is "food", the target work is a work of a travel landscape type, the similarity between the target work and the first work in the first work set is small, the characteristic difference value is greater than or equal to a preset threshold value, the target work is marked (marked as a second travel landscape) again, and the "travel landscape" is used as the label of the target work of which the characteristic difference value is greater than or equal to the preset threshold value.
In the above exemplary implementation, by labeling the target work whose feature difference is greater than or equal to the preset threshold value to obtain the second labeling information, and determining the second labeling information as the label of the target work whose feature difference is greater than or equal to the preset threshold value, after obtaining different types of user works, the label of the target work can be determined according to the relationship between the feature of the specific target work and the average feature of the first work in the first work set, thereby avoiding the phenomenon that the target work and the label are inapplicable due to direct determination of the user portrait as the label of the target work, ensuring the consistency of the content of the target work and the corresponding label, and meanwhile, when obtaining a work inconsistent with the user portrait label of the user, re-labeling is performed, and a new label is obtained, which can enrich the types of the user labels and obtain a more accurate label.
In one particular exemplary embodiment, user portrait determination and composition tag determination are included.
User portrait determination: for a user with K works, in any UGC (user generated Content) platform, the playing amount of the work corresponding to the user can be obtained, and the playing amount of each work is recorded as viewk(K is a positive integer) by calculating the average playsize view of the K pieces of workk avgScreening out the broadcast volume larger than viewk avgAnd using modeloriThe model extracts the features of the M works and records the extracted features as featuresmFor feature, using density clustering method (e.g. DBscan)mClustering is performed, and the cluster (or clusters) with the highest density is used as the representative works of the user, for example, N representative works are obtained, andthe N representative works are handed to a marking person, and the user is given a portrait label.
Product label determination: for any user with user portrait, obtaining the most representative N works as a feature extraction modeloriUsing a modeloriExtracting features from the N works and recording as featurei(i ═ 1,2,3, … …), the average avg _ feature is calculated by counting the N featuresj(j represents different representative product clusters), and the calculation method is shown as formula (1):
for a target work, a model is usedoriExtracting feature of the featurenewAnd calculates avg _ feature and featurenewThe cosine distance of (1) is calculated by the following formula (2):
distance＝cos(avg_feature，featurenew) (2)
finally, comparing the target product with a preset threshold (for example, the preset threshold is 0.2), determining whether the label of the target product corresponds to the user portrait label, as shown in formula (3):
wherein label is the label of the target work, Drop indicates that the label of the target work is other labels, and Neep indicates that the label of the target work corresponds to the user portrait label.
In the above exemplary embodiment, the method for determining the label of the target work according to the relationship between the specific target work features and the average features of the N works can avoid the phenomenon that the target work and the label are not applicable due to the fact that the user portrait is directly determined as the label of the target work, and ensure the consistency of the content of the target work and the corresponding label.
Fig. 5 is a block diagram illustrating a tag determination apparatus according to an example embodiment. Referring to fig. 5, the apparatus includes a user work set acquisition unit 501, a target work acquisition unit 502, a feature difference value determination unit 503, and a tag determination unit 504.
A user work set obtaining unit 501, configured to perform obtaining of a first work set corresponding to a user account, and extracting an average feature of a first work in the first work set; the first work in the first work set is a representative work corresponding to the user account;
a target work obtaining unit 502 configured to perform obtaining a target work and extracting features of the target work;
a feature difference determination unit 503 configured to perform calculating a feature difference between the features of the target work and the average features;
the tag determination unit 504 is configured to perform acquiring a user portrait tag corresponding to the user account, and determine the user portrait tag as a tag of the target work with a characteristic difference value smaller than a preset threshold value.
With regard to the apparatus in the above-described embodiment, the specific manner in which each unit performs the operation has been described in detail in the embodiment related to the method, and will not be described in detail here.
FIG. 6 is a block diagram illustrating an electronic device 600 for tag determination, according to an example embodiment. For example, the device 600 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.
Referring to fig. 6, device 600 may include one or more of the following components: a processing component 602, a memory 604, a power component 606, a multimedia component 608, an audio component 610, an interface to input/output (I/O) 612, a sensor component 614, and a communication component 616.
The processing component 602 generally controls overall operation of the device 600, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 602 may include one or more processors 620 to execute instructions to perform all or a portion of the steps of the methods described above. Further, the processing component 602 can include one or more modules that facilitate interaction between the processing component 602 and other components. For example, the processing component 602 can include a multimedia module to facilitate interaction between the multimedia component 608 and the processing component 602.
The memory 604 is configured to store various types of data to support operation at the device 600. Examples of such data include instructions for any application or method operating on device 600, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 604 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
A power supply component 606 provides power to the various components of the device 600. The power components 606 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 600.
The multimedia component 608 includes a screen that provides an output interface between the device 600 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 608 includes a front facing camera and/or a rear facing camera. The front camera and/or the rear camera may receive external multimedia data when the device 600 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.
The audio component 610 is configured to output and/or input audio signals. For example, the audio component 610 includes a Microphone (MIC) configured to receive external audio signals when the device 600 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal may further be stored in the memory 604 or transmitted via the communication component 616. In some embodiments, audio component 610 further includes a speaker for outputting audio signals.
The I/O interface 612 provides an interface between the processing component 602 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.
The sensor component 614 includes one or more sensors for providing status assessment of various aspects of the device 600. For example, the sensor component 614 may detect an open/closed state of the device 600, the relative positioning of components, such as a display and keypad of the device 600, the sensor component 614 may also detect a change in position of the device 600 or a component of the device 600, the presence or absence of user contact with the device 600, orientation or acceleration/deceleration of the device 600, and a change in temperature of the device 600. The sensor assembly 614 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 614 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 614 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component 616 is configured to facilitate communications between the device 600 and other devices in a wired or wireless manner. The device 600 may access a wireless network based on a communication standard, such as WiFi, an operator network (such as 2G, 3G, 4G, or 5G), or a combination thereof. In an exemplary embodiment, the communication component 616 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 616 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the device 600 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.
In an exemplary embodiment, a storage medium comprising instructions, such as the memory 604 comprising instructions, executable by the processor 620 of the device 600 to perform the above-described method is also provided. Alternatively, the storage medium may be a non-transitory computer readable storage medium, for example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
Fig. 7 is a block diagram illustrating an apparatus 700 for tag determination, according to an example embodiment. For example, the apparatus 700 may be provided as a server. Referring to fig. 7, apparatus 700 includes a processing component 722 that further includes one or more processors and memory resources, represented by memory 732, for storing instructions, such as applications, that are executable by processing component 722. The application programs stored in memory 732 may include one or more modules that each correspond to a set of instructions. Further, the processing component 722 is configured to execute instructions to perform the tag determination methods described above.
The apparatus 700 may also include a power component 726 configured to perform power management of the apparatus 700, a wired or wireless network interface 750 configured to connect the apparatus 700 to a network, and an input output (I/O) interface 750. The apparatus 700 may operate based on an operating system stored in memory 732, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, or the like.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (10)
1. A tag determination method, comprising:
acquiring a first work set corresponding to a user account, and extracting average characteristics of first works in the first work set; a first work in the first work set is a representative work corresponding to the user account;
obtaining a target work and extracting the characteristics of the target work;
calculating a feature difference between the features of the target work and the average features;
and acquiring a user portrait label corresponding to the user account, and determining the user portrait label as a label of the target work of which the characteristic difference value is smaller than a preset threshold value.
2. The tag determination method according to claim 1, wherein the obtaining of the first work set corresponding to the user account previously includes:
acquiring the first work set, and labeling each first work in the first work set to obtain first labeling information;
and determining the first labeling information as a user portrait label corresponding to the user account.
3. The tag determination method of claim 1, wherein the obtaining of the first set of works corresponding to the user account comprises:
acquiring a second work set corresponding to the user account; a second work in the second work set is a work corresponding to the user account;
screening the second work set according to the playing amount corresponding to each second work in the second work set to obtain a third work set;
extracting the characteristics of each third work in the third work set;
and clustering the characteristics of the third works, and screening the third work set according to clustering results to obtain the first work set.
4. The tag determination method according to claim 3, wherein the screening the second work set according to the playing amount corresponding to each second work in the second work set to obtain a third work set comprises:
calculating the average playing amount of the second work set according to the playing amount corresponding to each second work in the second work set;
and determining a set formed by the second works of which the playing amount is larger than the average playing amount as the third work set.
5. The tag determination method of claim 3, wherein the clustering the features of the third work and screening the third work set according to the clustering result to obtain the first work set comprises:
and clustering the characteristics of the third works, and determining a set formed by the third works with the highest density as the first work set.
6. The tag determination method of claim 1, wherein said extracting average features of a first work in the first work set comprises:
obtaining characteristics of each first work in the first work set;
calculating an average characteristic of the first work set according to the characteristic of each first work in the first work set.
7. The tag determination method according to claim 1, wherein the obtaining of the user portrait tag corresponding to the user account determines the user portrait tag as a tag of a target work of which the characteristic difference is smaller than a preset threshold, and then includes:
marking the target work with the characteristic difference value larger than or equal to the preset threshold value to obtain second marking information;
and determining the second labeling information as the label of the target work with the characteristic difference value larger than or equal to the preset threshold value.
8. A tag determination apparatus, comprising:
the system comprises a user work set acquisition unit, a first work set acquisition unit and a second work set acquisition unit, wherein the user work set acquisition unit is configured to acquire a first work set corresponding to a user account and extract average characteristics of first works in the first work set; a first work in the first work set is a representative work corresponding to the user account;
a target work acquisition unit configured to perform acquisition of a target work and extract a feature of the target work;
a feature difference determination unit configured to perform calculating a feature difference between the feature of the target work and the average feature;
and the label determining unit is configured to execute the steps of obtaining a user portrait label corresponding to the user account, and determining the user portrait label as a label of the target work of which the characteristic difference value is smaller than a preset threshold value.
9. An electronic device, comprising:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the tag determination method of any one of claims 1 to 7.
10. A storage medium having instructions that, when executed by a processor of the electronic device, enable the electronic device to perform the label determination method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010112968.9A CN111324755A (en) | 2020-02-24 | 2020-02-24 | Label determining method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010112968.9A CN111324755A (en) | 2020-02-24 | 2020-02-24 | Label determining method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111324755A true CN111324755A (en) | 2020-06-23 |
Family
ID=71171219
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010112968.9A Pending CN111324755A (en) | 2020-02-24 | 2020-02-24 | Label determining method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111324755A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112667906A (en) * | 2020-12-31 | 2021-04-16 | 上海众源网络有限公司 | Recommendation method and device for up master and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090228510A1 (en) * | 2008-03-04 | 2009-09-10 | Yahoo! Inc. | Generating congruous metadata for multimedia |
| US20130283307A1 (en) * | 2012-04-18 | 2013-10-24 | Narb Avedissian | System and methods for providing user generated video reviews |
| US20160328401A1 (en) * | 2015-05-05 | 2016-11-10 | Adobe Systems Incorporated | Method and apparatus for recommending hashtags |
| CN109947964A (en) * | 2019-04-02 | 2019-06-28 | 北京字节跳动网络技术有限公司 | Method and apparatus for more new information |
| CN110287375A (en) * | 2019-05-30 | 2019-09-27 | 北京百度网讯科技有限公司 | The determination method, apparatus and server of video tab |
-
2020
- 2020-02-24 CN CN202010112968.9A patent/CN111324755A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090228510A1 (en) * | 2008-03-04 | 2009-09-10 | Yahoo! Inc. | Generating congruous metadata for multimedia |
| US20130283307A1 (en) * | 2012-04-18 | 2013-10-24 | Narb Avedissian | System and methods for providing user generated video reviews |
| US20160328401A1 (en) * | 2015-05-05 | 2016-11-10 | Adobe Systems Incorporated | Method and apparatus for recommending hashtags |
| CN109947964A (en) * | 2019-04-02 | 2019-06-28 | 北京字节跳动网络技术有限公司 | Method and apparatus for more new information |
| CN110287375A (en) * | 2019-05-30 | 2019-09-27 | 北京百度网讯科技有限公司 | The determination method, apparatus and server of video tab |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112667906A (en) * | 2020-12-31 | 2021-04-16 | 上海众源网络有限公司 | Recommendation method and device for up master and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3179408B1 (en) | Picture processing method and apparatus, computer program and recording medium | |
| EP3327590B1 (en) | Method and device for adjusting video playback position | |
| CN105845124B (en) | Audio processing method and device | |
| CN106331761A (en) | Live broadcast list display method and apparatuses | |
| CN109819288B (en) | Method and device for determining advertisement delivery video, electronic equipment and storage medium | |
| CN107666536B (en) | Method and device for searching terminal | |
| CN107423386B (en) | Method and device for generating electronic card | |
| CN113099297B (en) | Method and device for generating click video, electronic equipment and storage medium | |
| CN109168062A (en) | Methods of exhibiting, device, terminal device and the storage medium of video playing | |
| CN109543066A (en) | Video recommendation method, device and computer readable storage medium | |
| CN107454204B (en) | User information labeling method and device | |
| CN106777016B (en) | Method and device for information recommendation based on instant messaging | |
| CN104636453A (en) | Illegal user data identification method and device | |
| CN109543069B (en) | Video recommendation method and device and computer-readable storage medium | |
| CN109670077A (en) | Video recommendation method, device and computer readable storage medium | |
| CN106547850B (en) | Expression annotation method and device | |
| CN112672208A (en) | Video playing method, device, electronic equipment, server and system | |
| CN105100193A (en) | Cloud business card recommendation method and device | |
| CN110019897B (en) | Method and device for displaying picture | |
| CN107801282A (en) | Desk lamp, desk lamp control method and device | |
| CN110941727A (en) | Resource recommendation method and device, electronic equipment and storage medium | |
| CN107454359B (en) | Method and device for playing video | |
| CN111526380B (en) | Video processing method, video processing device, server, electronic equipment and storage medium | |
| CN111324755A (en) | Label determining method and device, electronic equipment and storage medium | |
| CN109145151B (en) | Video emotion classification acquisition method and device |
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 |