CN111611973B - Target user identification method, device and storage medium - Google Patents
Target user identification method, device and storage medium Download PDFInfo
- Publication number
- CN111611973B CN111611973B CN202010486582.4A CN202010486582A CN111611973B CN 111611973 B CN111611973 B CN 111611973B CN 202010486582 A CN202010486582 A CN 202010486582A CN 111611973 B CN111611973 B CN 111611973B
- Authority
- CN
- China
- Prior art keywords
- user
- audio
- video
- quality
- detection model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000001514 detection method Methods 0.000 claims abstract description 128
- 238000012216 screening Methods 0.000 claims description 55
- 230000004044 response Effects 0.000 claims description 13
- 230000006399 behavior Effects 0.000 claims description 9
- 239000002131 composite material Substances 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000010801 machine learning Methods 0.000 abstract description 11
- 238000004519 manufacturing process Methods 0.000 description 38
- 230000000694 effects Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 8
- 238000013527 convolutional neural network Methods 0.000 description 6
- 238000002372 labelling Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 3
- 238000007418 data mining Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Databases & Information Systems (AREA)
- Human Computer Interaction (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Fuzzy Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Image Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The present disclosure discloses a method, a device and a storage medium for identifying a target user, firstly, audio and video contents issued by at least one user are obtained, each audio and video content is input into a video index detection model, and whether the at least one user is the target user is determined according to a detection result output by the model. Through the scheme, the audio and video content is input into the video index detection model to screen each user, so that when the user level to be detected is large, a high-quality original producer can be automatically identified and screened out as a target user by means of the machine learning model, the condition that the target user is missed due to manual annotation is avoided, and therefore the identification efficiency and accuracy of the high-quality UGC producer are improved.
Description
Technical Field
The disclosure relates to the technical field of machine learning, and in particular relates to a target user identification method, a target user identification device and a storage medium.
Background
With the increasing development of internet technology, websites or applications for uploading audio and video contents by users themselves are increasing, such audio and video contents uploaded by users themselves may be referred to as user originated contents (User Generated Content, UGC), and users uploading UGC may be referred to as UGC producers.
With the increasing number of UGC producers, how to screen out high quality UGC producers is an urgent issue to be resolved by the server of the UGC publishing platform. In the related art, for an application program providing a UGC service, in order to identify a high-quality UGC producer, a server of the application program may first classify uploading sources of content, then calculate according to posterior data such as the playing times of the content, the playing time of the content, the praise of a user, and the like, to obtain a candidate set, and finally screen the high-quality UGC producer from the candidate set through manual discovery and manual labeling.
However, the method for identifying the UGC producers by the scheme in the related art is easy to mine a large number of invalid accounts due to misjudging the uploading mode of the content, has strong dependence on manpower, and is easy to miss the high-quality UGC producers due to overlarge magnitude, so that the high-quality UGC producers are not high in identification efficiency and accuracy.
Disclosure of Invention
The disclosure provides a method, a device and a storage medium for identifying a target user, which can improve the identification efficiency and accuracy of the target user through multi-layer screening, and the technical scheme is as follows:
in one aspect, a method for identifying a target user is provided, which is characterized in that the method includes:
Acquiring each audio and video content issued by at least one user;
inputting the audio and video contents into a video index detection model to obtain a detection result of the video index detection model; the video index detection model comprises at least one of a quality detection model and a face recognition model, wherein the quality detection model is used for detecting quality indexes of all the audio and video contents, and the face recognition model is used for detecting whether the faces in all the audio and video contents are matched with the faces of the current user or not;
and determining a target user in the at least one user, wherein the target user is a user corresponding to the detection result meeting a specified condition.
In one aspect, there is provided an apparatus for identifying a target user, the apparatus comprising:
the content acquisition module is used for acquiring each audio and video content issued by at least one user;
the result acquisition module is used for inputting the audio and video contents into a video index detection model to obtain the detection result of the video index detection model; the video index detection model comprises at least one of a quality detection model and a face recognition model, wherein the quality detection model is used for detecting quality indexes of all the audio and video contents, and the face recognition model is used for detecting whether the faces in all the audio and video contents are matched with the faces of the current user or not;
And the target user determining module is used for determining a target user in the at least one user, wherein the target user is a user corresponding to the detection result and meeting a specified condition.
In yet another aspect, a computer device is provided that includes a processor and a memory storing at least one instruction, at least one program, code set, or instruction set that is loaded and executed by the processor to implement a method of target user identification as described in any of the alternative implementations above.
In another aspect, a computer readable storage medium having stored therein at least one instruction, at least one program, code set, or instruction set loaded and executed by a processor to implement a method of target user identification as described in any of the above alternative implementations is provided.
At least one aspect relates to a computer program product configured to cause: the method of the above aspect is performed by a data processing system comprising a processor and a memory, when executed on the data processing system. The computer program product may be embodied in or provided on a tangible, non-transitory computer readable medium.
The technical scheme provided by the embodiment of the disclosure can comprise the following beneficial effects:
according to the target user identification scheme provided by the embodiment of the disclosure, firstly, audio and video contents issued by at least one user are acquired, each audio and video content is input into a video index detection model, and whether the at least one user is a target user is determined according to a detection result which can be output by the model. Through the scheme, the audio and video content is input into the video detection model to screen each user, so that when the user level to be detected is large, a high-quality original producer can be automatically identified and screened out as a target user by means of the machine learning model, the condition that the target user is missed due to manual labeling is avoided, and therefore the identification efficiency and accuracy of the high-quality UGC producer are improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 is a schematic diagram of a target user identification system, according to an exemplary embodiment;
FIG. 2 is a flow chart illustrating a method of target user identification according to an exemplary embodiment;
FIG. 3 is a flow chart illustrating a method of target user identification according to another exemplary embodiment;
FIG. 4 is a schematic diagram of a method of target user identification according to the embodiment of FIG. 3;
FIG. 5 is a block diagram illustrating the structure of a target user identification apparatus according to an exemplary embodiment;
fig. 6 is a schematic diagram of a computer device, according to an example embodiment.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated.
FIG. 1 is a schematic diagram of a target user identification system, according to an exemplary embodiment. The target user identification system includes a platform server 110, a user terminal 120, and a database 130.
The user can enter a platform scene corresponding to the platform server 110 on the user terminal 120, and the user can upload original content of the user directly or by means of a third party platform under the platform scene.
After the user enters the platform scene, the platform server 110 may record the user data of the user in the platform scene as part of the basic information data.
The basic information data may include information data of uploading production content by a user in a scene, behavior information data of the user in the scene, consumption information data of the user, and the like.
The platform server 110 may include a memory that may be used to store the various user basic information data.
The platform server 110 may also send the collected user basic information data to the database 130 for storage. The database 130 may be a dedicated database for storing user basic information data collected by the platform, among other things.
The user terminal 120 may perform data transmission with the platform server 110 through a wired or wireless network.
The platform server 110 may invoke the user basic information data in the database 130 for processing, including, the platform server 110 may set up a machine learning model to mine the data.
The platform server 110 may be a server, or may be a server cluster formed by a plurality of servers, or may include one or more virtualized platforms, or may also be a cloud computing service center.
Fig. 2 is a flow chart illustrating a method of target user identification according to an exemplary embodiment. The method of target user identification may be performed by a computer device. The computer device may be the platform server 110 in the system shown in fig. 1. As shown in fig. 2, the method for identifying the target user may include the following steps:
in step 201, each audio/video content published by at least one user is acquired.
In the embodiment of the disclosure, the user original content release platform is a network platform for opening a user original content release function to a user, and on the user original content release platform, the user can release audio and video content and also can view the audio and video content released by other users. Correspondingly, the platform server corresponding to the original content release platform of the user can acquire each audio and video content released by at least one user in the platform.
The audio and video contents are audio and video files recorded and uploaded by a user through an application program or a third party application program of the platform, and other users of the platform can download each audio and video content through a network or watch each audio and video content online.
In addition, before each audio/video content published by at least one user in the user original content UGC publishing platform is obtained, at least one basic information data of each user can be judged according to a preliminary identification standard, and a respective preliminary identification result of each user is obtained, wherein the preliminary identification result is used for indicating whether the corresponding user passes the preliminary screening. And screening at least one user passing the preliminary screening from the users according to the respective preliminary identification results of the users.
In one possible implementation manner, various types of basic information data of the first user can be obtained from the database, index score conversion is respectively carried out on the various types of basic information data according to the primary identification standard to obtain index scores respectively corresponding to the various types of basic information data, then the comprehensive score of the first user is determined according to the weight parameters and the index scores respectively corresponding to the various types of basic information data, and finally the primary identification result of the first user is obtained according to the comprehensive score of the first user.
Wherein the first user may be any one of the respective users.
When the comprehensive score is greater than or equal to a first score threshold, determining that the primary identification result of the first user passes the primary screening; when the composite score is less than the first score threshold, the preliminary identification result of the first user may be determined as failing the preliminary screening.
In step 202, inputting each audio and video content into a video index detection model to obtain a detection result of the video index detection model; the video index detection model comprises at least one of a quality detection model and a face recognition model, wherein the quality detection model is used for detecting quality indexes of all audio and video contents, and the face recognition model is used for detecting whether faces in all audio and video contents are matched with faces of a current user or not.
Wherein the quality detection model may be a machine learning model for detecting audio quality as well as picture quality; the quality index may be an index value that measures audio quality as well as picture quality.
In the embodiment of the disclosure, in order to screen out high-quality original producers in each user as target users, a platform is convenient to focus on and cultivate the target users, and detection results can be obtained by detecting the quality of each aspect of original content of each user.
The detection of the original content of the user can be divided into quality detection of audio and video and originality detection of the audio and video.
The originality detection of the audio and video can be performed by judging whether the portrait appearing in the audio and video content of the current user corresponds to the portrait of the current user, if the portrait in the audio and video content is consistent with the portrait of the current user, the probability that the current user is an original content user is high.
In one possible implementation, a video index detection model may be used to detect audio and video quality and image quality as well as the consistency of the portraits in the audio and video with the portraits of the current user. The video index detection model can be divided into two models for detection in two aspects respectively.
The quality detection model can output the audio quality and picture quality conditions of all audio and video files by inputting all the audio and video files; the face recognition model can output the consistency degree between the face in each audio and video content and the face of the current user by inputting each audio and video file and the face information of the current user.
In one possible implementation, in response to the video index detection model including the quality detection model, each audio-video content may be input into the quality detection model to obtain an audio-video quality score for the each audio-video content.
The platform server can detect each audio and video content through the quality detection model to obtain audio and video quality information of each audio and video content, then obtain quality standard rate of each audio and video content according to the audio and video quality information of each audio and video content, and the platform server can take the quality standard rate of each audio and video content as the audio and video quality score of each audio and video content.
The quality standard reaching rate can be used for indicating the proportion of the corresponding audio and video quality information reaching the quality index in each audio and video content.
In another possible implementation manner, in response to the video index detection model including a face recognition model, the platform server may input image information corresponding to each audio and video content and a face image of a corresponding user into the face recognition model, so as to obtain a face matching rate output by the face recognition model.
The face matching rate can be used for indicating the proportion of the audio and video contents which are matched with the face image aiming at the image information corresponding to any user in each audio and video content.
The face matching rate may also be used to indicate a matching degree between image information of each audio/video content corresponding to at least one user and a face image of the corresponding user.
In step 203, a target user among the at least one user is determined, the target user being used to indicate a user whose detection result satisfies a specified condition.
In the embodiment of the present disclosure, the detection results obtained through the foregoing steps, including the audio-visual quality detection result and the detection result of the face consistency, may compare the detection result with the specified condition, and determine that the detection result of the at least one user meets the specified condition, thereby determining that the at least one user is the target user.
In one possible implementation manner, a user with the audio-video quality score higher than the second score threshold value of the corresponding audio-video content is determined as a target user.
The second score threshold may be used to indicate a quality achievement rate indicator for the audio-video content determined to be the target user.
In another possible implementation manner, it is determined that, among at least one user, a user whose corresponding face matching rate is higher than a matching rate threshold is a target user.
In summary, according to the target user identification scheme provided by the embodiment of the present disclosure, firstly, audio and video contents published by at least one user are acquired, each audio and video content is input into a video index detection model, and whether the at least one user is a target user is determined according to a detection result output by the model. Through the scheme, the audio and video content is input into the video detection model to screen each user, so that when the user level to be detected is large, a high-quality original producer can be automatically identified and screened out as a target user by means of the machine learning model, the condition that the target user is missed due to manual labeling is avoided, and therefore the identification efficiency and accuracy of the high-quality UGC producer are improved.
Fig. 3 is a flow chart illustrating a method of target user identification according to another exemplary embodiment. The method of target user identification may be performed by a computer device. The computer device may be the platform server 110 in the system shown in fig. 1. As shown in fig. 3, the method for identifying the target user may include the following steps:
in step 301, various types of basic information data of a first user are acquired from a database.
In the embodiment of the disclosure, the basic information data corresponding to any one of the users is obtained from the database.
In one possible implementation manner, the basic information data corresponding to the user may be at least one of user behavior information data corresponding to the user and account consumption information data corresponding to the user.
The user behavior information data may be used to indicate user operation data and time data corresponding to when the user issues audio and video content (such as UGC).
In one possible implementation, the user behavior information data includes at least one of activity of the production content corresponding to the user, production source of the production content uploaded by the user, number of fans of the user account corresponding to the user, and fan situation of fan quality.
For example, when the user behavior information data is the activity level of the production content corresponding to the user, the production content may be used to indicate the audio/video content released by the user on the content release platform. For example, the liveness of the production content corresponding to the user is expressed as the number of times the user uploads the production content to the platform server or the data amount of the uploaded production content in a specified period of time.
When the user behavior information data is a production source from which the user uploads the production content, the production source from which the user uploads the production content is represented as content produced by what channel the user has produced and production source information data from which the production content has been uploaded in a direct or indirect manner.
When the user behavior information data are the number of the fans of the user account corresponding to the user and the fan situation of the fan quality, the number of the fans of the user account corresponding to the user and the fan situation of the fan quality are expressed as the number of the fans actually owned in the user account corresponding to the user and the activity degree of each fan.
The quality of the fan can be judged by detecting the times of opening the user's production content and the total watching duration or the average duration of all users focusing on the user; for example, when the average time period for all users who pay attention to the users to watch the production content of the users exceeds a specified threshold, it may be determined that the fan quality of the users meets the average standard.
For example, the account consumption information data corresponding to the user may be consumption information data of an account corresponding to the user, and the consumption information data of the account corresponding to the user may include historical consumption information data of a user account corresponding to the user, platform member class information data, and the like.
In one possible implementation, the database may be a database used in the platform server to collect the basic information data corresponding to the user, or a database used in other storage devices outside the platform server to store the basic information data.
In one possible implementation manner, the platform server may retrieve each basic information data in the database corresponding to the user according to the user account corresponding to the user, or may classify and retrieve each user in the database according to the kind of the basic information data, so as to obtain the data of each user corresponding to each kind of the basic information data.
In step 302, index score conversion is performed on each type of basic information data according to the preliminary identification criteria, so as to obtain index scores corresponding to each type of basic information data.
In the embodiment of the disclosure, index scores of corresponding categories are obtained according to the types of the basic information data according to the obtained basic information data of each user. And comparing and judging various basic information data of each user through the index score to obtain a judging result corresponding to each basic information data of each user.
Wherein the index score may be used to indicate that the underlying information data reaches a score corresponding to the index that passed the preliminary screening.
In one possible implementation, the index scores of the basic information data of each kind may be calculated according to data obtained from a database in real time.
For example, production activity information data of a plurality of users are obtained from a database, an average value of the number of times of uploading production content by the plurality of users within a specified time limit, such as 30 days, is calculated, when the average value of the number of production content by the plurality of users within thirty days is calculated to be 30, it can be determined that an index score corresponding to the production activity information data can be 30, and a judgment result corresponding to each basic information data of the plurality of users can be obtained by comparing the production activity information data of each user with the index score, that is, a user judgment result that the number of production content exceeds 30 in thirty days is standard, and a user judgment result that the number of production content does not exceed 30 in thirty days is not standard.
In one possible implementation manner, the actual basic information data corresponding to each user is compared with the calculated index score to obtain a data value different from the index score or a data value exceeding the index score, and the score of the single basic information data corresponding to each user is obtained according to the data value.
For example, for the production activity information data of the user, according to the actual attribute of the user release platform, the index score of the production activity information data of the user can be specified to be 60 scores of thirty pieces of production content corresponding to the basic information data within thirty days, the score can be specified to be reduced by 3 scores when the data amount of the production content within thirty days of the user is 1 piece, the score can be increased by 3 scores when the data amount of the production content exceeds 1 piece, and the score of the production activity information data of the user can be obtained according to the specified rule.
In step 303, a composite score of the first user is determined according to the weight parameters and the index scores respectively corresponding to the various types of basic information data.
In the embodiment of the disclosure, according to index scores of basic information data of each category corresponding to any one of the users, a comprehensive score of any one of the users is determined by combining weight parameters corresponding to the basic information data of each category.
The weight parameters can be preset according to the attribute of the UGC release platform where the user is located, and different weight parameters can be set for each kind of basic information data.
The attribute of the UGC publishing platform can be used for indicating analysis emphasis conditions of the UGC publishing platform on various basic information data.
In one possible implementation, the corresponding weight parameters are set according to the size of the attribute of the product platform, which affects the UGC publication platform corresponding to the various basic information data.
For example, in a UGC publishing platform, the platform needs to acquire a user with higher value and heat provided for the platform, so that the weight parameters occupied by consumption condition information data and vermicelli condition information data can be properly improved, the dependence on the two basic information data is higher in the calculation process, and the user meeting the standard is more accurately acquired.
In one possible implementation manner, the score corresponding to each kind of basic information data is weighted and calculated through preset weights corresponding to each kind of basic information data, so as to obtain the comprehensive score corresponding to each user.
For example, when the platform server of a platform presets two kinds of basic information data to be calculated, the basic information data are consumption condition information data and vermicelli condition information data respectively. And when index score conversion is respectively carried out on the two types of basic information data according to the primary identification standard, the score corresponding to the consumption condition information data of the current user is 50, the score corresponding to the vermicelli condition information data is 60, and the weight parameters respectively set by the consumption condition information data and the vermicelli condition information data in the platform are 0.3 and 0.7, the comprehensive score corresponding to the current user is calculated to be 50 x 0.3+60 x 0.7.
In step 304, a preliminary identification result of the first user is obtained according to the composite score of the first user.
In the embodiment of the disclosure, the platform server may determine the preliminary identification result corresponding to each user according to the comprehensive score of any one of the users.
In one possible implementation, the preliminary identification result of the first user is determined to pass the preliminary screening in response to the composite score being greater than or equal to a first score threshold, and the preliminary identification result of the first user is determined to fail the preliminary screening in response to the composite score being less than the first score threshold.
The data information of the user passing the preliminary screening can be continuously acquired and processed, and the data information of the user not passing the preliminary screening can be directly discarded, namely, the subsequent processing is not performed.
In step 305, at least one user passing the preliminary screening is screened from the respective users according to the respective preliminary identification results of the respective users.
In the embodiment of the disclosure, each audio and video content issued by the preliminarily screened user in the platform server is acquired, the next detection of the audio and video content is performed, and each user is further screened.
The initial target user is used for indicating the initial recognition result to indicate the user passing the initial screening.
In step 306, each audio/video content issued by at least one user is obtained, and each audio/video content is input into the quality detection model in response to the video index detection model including the quality detection model, so as to obtain the audio/video quality score of each audio/video content.
In the embodiment of the disclosure, audio and video contents corresponding to at least one user through preliminary screening are input into a quality detection model, and audio and video quality scores corresponding to the audio and video contents are obtained.
The audio-video content comprises at least one of audio content and image content.
In one possible implementation manner, audio and video contents uploaded to the platform server by all users passing the preliminary screening are acquired and input into a quality detection model for detecting audio quality and picture quality.
The quality detection model may be an image segmentation model, a deep learning model in the visual field, for example, a machine learning model using convolutional neural networks (Convolutional Neural Networks, CNN).
For example, when the production content of the user subjected to the preliminary screening is obtained as short videos, each short video corresponding to the user can be respectively input into a quality detection model established in the computer equipment, and the score corresponding to each short video of each user and the score representing the condition of the picture image can be output as the audio-video quality score through the quality detection model.
In one possible implementation manner, the audio and video content of the user subjected to preliminary screening is input into the quality detection model, so that the detection score corresponding to each audio and video content is obtained as the audio and video quality score.
In one possible implementation, the audio-video content corresponding to the preliminarily filtered user may be each audio-video content corresponding to the user or each audio-video content uploaded within a specified period.
When the audio/video content is each audio/video content uploaded by the user correspondingly or each audio/video content uploaded within a specified period, the following two steps can be performed respectively.
1) When the audio and video contents corresponding to the user after the primary screening are used for indicating the audio and video contents corresponding to the user to upload, detecting the audio and video contents uploaded by the user after the primary screening through a quality detection model, and outputting the total standard reaching rate as the audio and video quality score.
The overall standard reaching rate can be used for indicating the proportion condition of the audio-video content detected to reach the standard in all the audio-video content.
2) When the audio and video content corresponding to the user after the primary screening is used for indicating the audio and video content uploaded in the appointed time limit, the audio and video content uploaded by the user after the primary screening in the appointed time limit is detected through a quality detection model, and the recent standard reaching rate is output as the audio and video quality score.
The recent standard reaching rate can be used for indicating the occupation ratio condition of the audio-video content which is detected to reach the standard in the audio-video content uploaded within a specified period.
For example, short videos corresponding to users to be detected are input into a quality detection model for detection, the short videos can be divided into a sound part and an image picture part, the quality detection model can respectively detect the quality of sound and the image picture quality, whether the short videos reach standards in terms of sound and picture quality or not can be obtained by detecting the frequency of sound, the resolution frame number of pictures and the like as parameters, after all detection content resources of the users are detected, the number of short videos reaching standards through detection can be obtained, the number of short videos reaching standards through detection is compared with the number of short videos detected by the users, the duty ratio condition of the short videos reaching standards is obtained, and the platform server can obtain the users obtained through further screening according to the duty ratio condition.
In one possible implementation manner, the audio and video quality scores of the users are ranked to obtain an average value, and the corresponding users with the audio and video quality scores higher than the average value are determined as the users screened by audio and video quality detection.
In another possible implementation manner, the audio and video quality scores of the users are compared and judged according to preset indexes of the preset audio and video quality scores, and the users with the audio and video quality scores higher than the preset audio and video quality score indexes are determined as the users screened through audio and video quality detection.
In step 307, in response to the video index detection model including the face recognition model, the image information corresponding to each audio and video content and the face image corresponding to the user are input into the face recognition model, so as to obtain the face matching rate output by the face recognition model.
In the embodiment of the disclosure, audio and video content corresponding to at least one user which meets the standard through primary screening and actual face image information corresponding to each user are input into a face recognition model.
In one possible implementation, the actual face image information corresponding to each user may be retrieved from a memory or database of the platform server. The audiovisual content may include image content in the production content of the user.
The face recognition model may be an image segmentation model, a deep learning model in the visual field, for example, a machine learning model using convolutional neural networks (Convolutional Neural Networks, CNN).
In one possible implementation manner, face matching rates corresponding to all the sections of audio and video contents corresponding to all the users and face matching rates corresponding to the whole audio and video contents corresponding to all the users are obtained through a face recognition model to serve as recognition scores.
The identification score may include, among other things, an overall or recent rate of agreement.
In one possible implementation manner, the overall consistency rate may be obtained by inputting image portions in each audio-video content of the user into a face recognition model to obtain the number of audio-video contents passing through face recognition, and comparing the number of audio-video contents passing through face recognition with the number of overall audio-video contents to obtain the overall consistency rate as the recognition score.
In another possible implementation manner, the recent consistency rate may be obtained by inputting image portions in the audio and video contents corresponding to the user within a specified period into a face recognition model to obtain the number of audio and video contents passing the face recognition, and comparing the number of audio and video contents passing the face recognition with the number of detected audio and video contents to obtain the recent consistency rate as the recognition score.
In step 308, a target user of the at least one user is determined.
In the embodiment of the disclosure, at least one of a quality detection model and a face recognition model is used for detecting audio and video contents corresponding to the user after the primary screening, and a target user meeting specified conditions is obtained according to a detection result.
Wherein the target user may be used to indicate a user whose detection result satisfies a specified condition.
In one possible implementation, the user detected by the model is determined as the target user according to the score index set in advance or determined by calculation.
For example, the recognition scores or the audio and video quality scores of the users are ranked, the recognition scores which are obtained by training according with the screening attribute are selected, or an average value is obtained, or a designated threshold value is taken as a score index, so that the recognition scores are obtained, or the users with the audio and video quality scores higher than the score index correspond to the score index are used as target users screened through the model.
In one possible implementation, the user through the first screening may perform the subsequent screening through at least one of a quality detection model and a face recognition model to determine the target user.
After the user passes the primary screening, the screening of two machine learning models, namely a quality detection model and a face recognition model, can be performed.
In one possible implementation manner, the first-screened user is screened twice through the audio-video quality model, and then screened three times through the face recognition model, so as to determine the target user.
For example, fig. 4 is a schematic diagram of a method for identifying a target user according to an embodiment of the disclosure. As shown in fig. 4, in a real scene, in order to automatically screen all original contents of users, data mining may be performed from three dimensions of data index, audio and video quality detection and face image recognition. The target user identification method comprises the following steps:
in step 401, a task of identifying a target user is started, and various types of basic information data of respective users required for identifying the target user are acquired by querying a database.
In step 402, basic information data of various users including consumption information data, fan situation, activity of production content and production source are obtained, and whether the basic information data of each user meets the preliminary identification standard is judged according to the preliminary identification standard.
Through the first dimension, each user can be subjected to preliminary screening according to whether the basic information data of each user meets the preliminary identification standard or not, so that the users meeting the preliminary identification standard and the users not meeting the preliminary identification standard are obtained.
In step 403, the production content of the user meeting the preliminary identification criteria is input into a machine model for detecting the quality of the picture and the audio, the audio and image quality of the production content of the user is output by the quality detection model, and the user whose audio and image quality reaches the index is discovered for the next screening.
And (3) performing secondary screening on the users subjected to primary screening through the second dimension and the quality detection model, so that the users with the sound and image quality reaching the index are subjected to subsequent screening.
In step 404, the production content of the user whose audio and video quality has been detected to reach the index and the actual portrait image in the user information of the user are input into the face recognition model, the consistency degree of the production content and the face image of the user is determined, and the user whose consistency degree reaches the index is screened out.
And screening the producer for three times through the face recognition model in the third dimension to obtain the user with the consistency degree reaching the index, outputting the final detection result, and finishing the data mining of the target user.
In another possible implementation manner, the computer device may perform secondary screening on the audio and video content corresponding to the user that passes the primary screening through the face recognition model, and then perform tertiary screening through the quality detection model, so as to determine the target user.
For example, the audio and video content corresponding to the user after the primary screening and the actual portrait image in the user information are input into a face recognition machine model, the consistency degree of the audio and video content and the user portrait image is determined, and the user with the consistency degree reaching the standard is screened. And (3) carrying out secondary screening on the users through the second dimension face recognition model to obtain users with consistency degree reaching the index, inputting the audio and video contents of the users subjected to secondary screening into the quality detection model, outputting the audio and video quality conditions of the audio and video contents of the users, and finding out the users with the audio and video quality conditions reaching the index as target users. And thirdly screening the users through a third dimension and a quality detection model to obtain users with sound and image quality meeting the standard, and completing data mining of target users.
In addition, the users who pass the primary screening can be screened by the quality detection model and the face recognition model at the same time, and the users who pass the quality detection model and the face recognition model at the same time can be determined as target users by comparing and determining the users who pass the quality detection model and the face recognition model at the same time.
For example, audio and video contents corresponding to the users passing through the primary screening are simultaneously input into two models of a quality detection model and a face recognition model, the quality detection model and the face recognition model respectively identify and detect the audio and video contents corresponding to the users passing through the primary screening, users reaching standards through the quality detection model and users reaching standards through the face recognition model can be respectively obtained, the two users reaching standards through the face recognition model are compared, and the users reaching standards together are determined to be target users.
In summary, according to the target user identification scheme provided by the embodiment of the present disclosure, firstly, audio and video contents published by at least one user are acquired, each audio and video content is input into a video index detection model, and whether the at least one user is a target user is determined according to a detection result output by the model. Through the scheme, the audio and video content is input into the video detection model to screen each user, so that when the user level to be detected is large, a high-quality original producer can be automatically identified and screened out as a target user by means of the machine learning model, the condition that the target user is missed due to manual labeling is avoided, and therefore the identification efficiency and accuracy of the high-quality UGC producer are improved.
Fig. 5 is a block diagram illustrating a structure of a target user recognition apparatus according to an exemplary embodiment. The target user identifying means may be implemented as all or part of the server by means of hardware or a combination of hardware and software to perform all or part of the steps of the method shown in the corresponding embodiment of fig. 2 or fig. 3. The target user identification apparatus may include:
the content acquisition module 510 is configured to acquire each audio/video content published by at least one user;
the result obtaining module 520 is configured to input the audio and video contents into a video index detection model, and obtain a detection result of the video index detection model; the video index detection model comprises at least one of a quality detection model and a face recognition model, wherein the quality detection model is used for detecting quality indexes of all the audio and video contents, and the face recognition model is used for detecting whether the faces in all the audio and video contents are matched with the faces of the current user or not;
a target user determining module 530, configured to determine a target user among the at least one user, where the target user is a user whose corresponding detection result meets a specified condition.
In one possible implementation, in response to the video index detection model including the quality detection model, the apparatus includes:
the score acquisition module is used for inputting the audio and video contents into the quality detection model to acquire the audio and video quality scores of the audio and video contents;
and the first target determining module is used for determining that the user with the audio and video quality score higher than the second score threshold value in the corresponding audio and video content in the at least one user is the target user.
In one possible implementation manner, the score acquisition module includes:
the quality information acquisition sub-module is used for detecting each audio and video content through the quality detection model to obtain the audio and video quality information of each audio and video content;
the quality standard rate obtaining sub-module is used for obtaining the quality standard rate of each audio-video content according to the audio-video quality information of each audio-video content, and the quality standard rate is used for indicating the proportion of the audio-video content, of which the corresponding audio-video quality information reaches a quality index, in each audio-video content;
and the score determining sub-module is used for taking the quality standard reaching rate of each audio and video content as the audio and video quality score of each audio and video content.
In one possible implementation, in response to the video index detection model including the face recognition model, the apparatus includes:
the matching rate acquisition module is used for inputting the image information corresponding to each audio and video content and the face image of the corresponding user into the face recognition model to obtain the face matching rate output by the face recognition model, wherein the face matching rate is used for indicating the matching degree of the image information of each audio and video content and the face image of the corresponding user;
and the second target determining module is used for determining that the user with the corresponding face matching rate higher than a matching rate threshold value in the at least one user is the target user.
In one possible implementation, the apparatus further includes:
the primary result acquisition module is used for judging at least one basic information data of each user according to a primary identification standard to acquire a primary identification result of each user, wherein the primary identification result is used for indicating whether the corresponding user passes primary screening or not;
and the content acquisition sub-module is used for screening the at least one user which passes the preliminary screening from the users according to the respective preliminary identification results of the users.
In one possible implementation manner, the preliminary result obtaining module includes:
the data acquisition sub-module is used for acquiring various types of basic information data of the first user from the database; the first user is any one of the users;
the index score acquisition sub-module is used for respectively carrying out index score conversion on the various types of basic information data according to the preliminary identification standard to obtain index scores respectively corresponding to the various types of basic information data;
the comprehensive score acquisition sub-module is used for determining the comprehensive score of the first user according to the weight parameters and the index scores respectively corresponding to the various types of basic information data;
and the preliminary result acquisition sub-module is used for acquiring a preliminary identification result of the first user according to the comprehensive score of the first user.
In one possible implementation manner, the preliminary result obtaining sub-module includes:
the first result acquisition unit is used for determining that the primary identification result of the first user passes the primary screening in response to the comprehensive score being greater than or equal to a first score threshold;
and the second result acquisition unit is used for responding to the comprehensive score being smaller than the first score threshold value and determining that the primary identification result of the first user does not pass the primary screening.
In one possible implementation, the basic information data includes at least one of consumption information data and user behavior information data.
In summary, according to the target user identification scheme provided by the embodiment of the present disclosure, firstly, audio and video contents published by at least one user are acquired, each audio and video content is input into a video index detection model, and whether the at least one user is a target user is determined according to a detection result output by the model. Through the scheme, the audio and video content is input into the video detection model to screen each user, so that when the user level to be detected is large, a high-quality original producer can be automatically identified and screened out as a target user by means of the machine learning model, the condition that the target user is missed due to manual labeling is avoided, and therefore the identification efficiency and accuracy of the high-quality UGC producer are improved.
Fig. 6 is a schematic diagram of a computer device, according to an example embodiment. The computer apparatus 600 includes a central processing unit (Central Processing Unit, CPU) 601, a system Memory 604 including a random access Memory (Random Access Memory, RAM) 602 and a Read-Only Memory (ROM) 603, and a system bus 605 connecting the system Memory 604 and the central processing unit 601. The computer device 600 also includes a basic Input/Output system (I/O) 606 for facilitating the transfer of information between the various devices within the computer device, and a mass storage device 607 for storing an operating system 608, application programs 609, and other program modules 610.
The memory further includes one or more programs, one or more programs being stored in the memory, and the central processing unit 601 implements all or part of the steps of the method shown in fig. 2 or 3 by executing the one or more programs.
The disclosed embodiments also provide a computer device storage medium for storing computer device software instructions for use with the above-described test apparatus, which contains a program designed to perform the above-described target user identification method.
The disclosed embodiments also provide a computer program product storing at least one instruction that is loaded and executed by a processor to implement all or part of the steps of the methods described above with respect to the corresponding embodiments of fig. 2 or 3.
It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof.
Claims (11)
1. A method of target user identification, the method comprising:
acquiring each audio and video content issued by at least one user;
inputting the audio and video contents into a video index detection model to obtain a detection result of the video index detection model; the video index detection model comprises at least one of a quality detection model and a face recognition model, wherein the quality detection model is used for detecting quality indexes of all the audio and video contents, and the face recognition model is used for detecting whether the faces in all the audio and video contents are matched with the faces of the current user or not;
And determining a target user in the at least one user, wherein the target user is a user corresponding to the detection result meeting a specified condition.
2. The method of claim 1, wherein in response to the video index detection model comprising the quality detection model, the method comprises:
inputting the audio and video contents into the quality detection model to obtain the audio and video quality scores of the audio and video contents;
and determining that the user with the audio and video quality score higher than a second score threshold value in the corresponding audio and video content in the at least one user is the target user.
3. The method of claim 2, wherein said inputting the respective audio-visual contents into the quality detection model to obtain the audio-visual quality score of the respective audio-visual contents comprises:
detecting each audio and video content through the quality detection model to obtain audio and video quality information of each audio and video content;
acquiring the quality standard reaching rate of each audio-video content according to the audio-video quality information of each audio-video content, wherein the quality standard reaching rate is used for indicating the proportion of the audio-video content, of which the corresponding audio-video quality information reaches a quality index, in each audio-video content;
And taking the quality standard reaching rate of each audio and video content as the audio and video quality score of each audio and video content.
4. The method of claim 1, wherein in response to the video index detection model comprising the face recognition model, the method comprises:
inputting the image information corresponding to each audio and video content and the face image of the corresponding user into the face recognition model to obtain the face matching rate output by the face recognition model, wherein the face matching rate is used for indicating the matching degree of the image information of each audio and video content and the face image of the corresponding user;
and determining that the user with the corresponding face matching rate higher than a matching rate threshold value is the target user in the at least one user.
5. The method of claim 1, further comprising, prior to the obtaining each audio-visual content published by the at least one user:
judging at least one basic information data of each user according to a preliminary identification standard, and obtaining a preliminary identification result of each user, wherein the preliminary identification result is used for indicating whether the corresponding user passes preliminary screening;
And screening the at least one user which passes the preliminary screening from the users according to the respective preliminary identification results of the users.
6. The method according to claim 5, wherein the determining at least one basic information data of each user according to the preliminary identification criteria to obtain the respective preliminary identification result of each user includes:
acquiring various types of basic information data of a first user from a database; the first user is any one of the users;
according to the preliminary identification standard, respectively performing index score conversion on the various types of basic information data to obtain index scores respectively corresponding to the various types of basic information data;
determining the comprehensive score of the first user according to the weight parameters and the index scores respectively corresponding to the various types of basic information data;
and obtaining a preliminary identification result of the first user according to the comprehensive score of the first user.
7. The method of claim 6, wherein obtaining the preliminary identification result of the first user based on the composite score of the first user comprises:
Determining that the primary identification result of the first user passes primary screening in response to the composite score being greater than or equal to a first score threshold;
and determining that the primary identification result of the first user does not pass the primary screening in response to the integrated score being less than the first score threshold.
8. The method of claim 6, wherein the base information data comprises at least one of consumption information data and user behavior information data.
9. An apparatus for target user identification, the apparatus comprising:
the content acquisition module is used for acquiring each audio and video content issued by at least one user;
the result acquisition module is used for inputting the audio and video contents into a video index detection model to obtain the detection result of the video index detection model; the video index detection model comprises at least one of a quality detection model and a face recognition model, wherein the quality detection model is used for detecting quality indexes of all the audio and video contents, and the face recognition model is used for detecting whether the faces in all the audio and video contents are matched with the faces of the current user or not;
And the target user determining module is used for determining a target user in the at least one user, wherein the target user is a user corresponding to the detection result and meeting a specified condition.
10. A computer device comprising a processor and a memory having stored therein at least one instruction, at least one program, code set or instruction set, the at least one instruction, the at least one program, code set or instruction set being loaded and executed by the processor to implement the method of target user identification of any of claims 1 to 8.
11. A computer readable storage medium having stored therein at least one instruction, at least one program, code set, or instruction set, the at least one instruction, the at least one program, the code set, or instruction set being loaded and executed by a processor to implement the method of target user identification of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010486582.4A CN111611973B (en) | 2020-06-01 | 2020-06-01 | Target user identification method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010486582.4A CN111611973B (en) | 2020-06-01 | 2020-06-01 | Target user identification method, device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111611973A CN111611973A (en) | 2020-09-01 |
| CN111611973B true CN111611973B (en) | 2023-05-12 |
Family
ID=72197885
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010486582.4A Active CN111611973B (en) | 2020-06-01 | 2020-06-01 | Target user identification method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111611973B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113766306B (en) * | 2021-04-21 | 2023-11-14 | 腾讯科技（北京）有限公司 | Method, device, computer equipment and storage medium for detecting video clamping |
| CN114429605B (en) * | 2022-01-25 | 2024-07-02 | 南京杰迈视讯科技有限公司 | Dynamic random target identification, positioning and tracking system in wide area environment |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104966006A (en) * | 2015-05-14 | 2015-10-07 | 苏州市公安局苏州工业园区分局 | Intelligent face identification system based on cloud variation platform |
| CN109922334B (en) * | 2017-12-13 | 2021-11-19 | 阿里巴巴（中国）有限公司 | Video quality identification method and system |
| CN108985178A (en) * | 2018-06-21 | 2018-12-11 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating information |
| CN110110084A (en) * | 2019-04-23 | 2019-08-09 | 北京科技大学 | The recognition methods of high quality user-generated content |
| CN110309421B (en) * | 2019-06-20 | 2021-10-22 | 北京奇艺世纪科技有限公司 | UGC content quality evaluation method and device and electronic equipment |
| CN111159572B (en) * | 2019-12-18 | 2023-05-23 | 北京百度网讯科技有限公司 | Recommended content auditing method and device, electronic equipment and storage medium |
-
2020
- 2020-06-01 CN CN202010486582.4A patent/CN111611973B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111611973A (en) | 2020-09-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11055516B2 (en) | Behavior prediction method, behavior prediction system, and non-transitory recording medium | |
| CN108810642B (en) | Bullet screen display method and device and electronic equipment | |
| JP2021532434A (en) | Face feature extraction model Training method, face feature extraction method, device, equipment and storage medium | |
| CN109508406B (en) | Information processing method and device and computer readable storage medium | |
| CN111611973B (en) | Target user identification method, device and storage medium | |
| CN111464819B (en) | Live image detection method, device, equipment and storage medium | |
| CN111090813B (en) | Content processing method and device and computer readable storage medium | |
| CN109491915B (en) | Data processing method and device, medium and computing equipment | |
| CN109471981B (en) | Comment information sorting method and device, server and storage medium | |
| CN110895568B (en) | Method and system for processing court trial records | |
| CN112995690B (en) | Live content category identification method, device, electronic equipment and readable storage medium | |
| CN112232276B (en) | Emotion detection method and device based on voice recognition and image recognition | |
| CN113763348A (en) | Image quality determination method and device, electronic equipment and storage medium | |
| CN113762107A (en) | Object state evaluation method and device, electronic equipment and readable storage medium | |
| CN112330585A (en) | Image quality detection method and device and electronic equipment | |
| CN114245185B (en) | Video recommendation method, model training method, device, electronic equipment and medium | |
| CN113128522B (en) | Target identification method, device, computer equipment and storage medium | |
| CN113822907A (en) | Image processing method and device | |
| CN111325198B (en) | Video object feature extraction method and device, and video object matching method and device | |
| CN115983499A (en) | Box office prediction method and device, electronic equipment and storage medium | |
| CN110163043B (en) | Face detection method, device, storage medium and electronic device | |
| CN111291259B (en) | Data screening method and device, electronic equipment and storage medium | |
| CN113420809A (en) | Video quality evaluation method and device and electronic equipment | |
| CN113066024A (en) | Training method of image blur detection model, image blur detection method and device | |
| CN113242470A (en) | Video publishing method and device applied to foreign trade marketing |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231007 Address after: 31a, 15 / F, building 30, maple mall, bangrang Road, Brazil, Singapore Patentee after: Baiguoyuan Technology (Singapore) Co.,Ltd. Address before: 511402 5-13 / F, West Tower, building C, 274 Xingtai Road, Shiqiao street, Panyu District, Guangzhou City, Guangdong Province Patentee before: GUANGZHOU BAIGUOYUAN INFORMATION TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |