A video tag generation method, apparatus, device and medium

Abstract

The application discloses a video label generation method, device, equipment and medium, wherein the method comprises: obtaining related information corresponding to any to-be-processed video, the related information comprising image, audio, text information and the like of the to-be-processed video. The related information is input into a prediction model to obtain a prediction feature vector output by the prediction model. The prediction model is generated based on a video recommendation model and target training samples, wherein the video recommendation model is used to recommend videos to users, and can extract a feature vector reflecting the interest of the user from the video when recommending. That is, the prediction feature vector output by the prediction model can reflect the interest characteristics of the user. The prediction feature vector is clustered, and the label of the to-be-processed video is determined according to the clustering result.

Description

Technical Field

[0001] This application relates to the field of computer technology, specifically to a video tag generation method, apparatus, device, and medium. Background Technology

[0002] Video tagging is a common method for describing video content. Typically, video tag generation is based on a tagging system built using methods such as expert definition and operational data mining, with content recognition models determining the video's classification information. However, for videos with limited interaction, relying solely on expert-defined tagging systems often yields inaccurate tags. Therefore, how to match tags to recommended content is a pressing technical problem that needs to be solved. Summary of the Invention

[0003] In view of this, embodiments of this application provide a video tag generation method, apparatus, device, and medium to generate matching tags for videos, ensuring that the video tags are consistent with user interests.

[0004] To achieve the above objectives, the technical solution provided in this application is as follows:

[0005] In a first aspect of this application, a video tag generation method is provided, the method comprising:

[0006] Obtain relevant information about the video to be processed, including image, audio, and text information of the video to be processed;

[0007] The relevant information is input into the prediction model to obtain the prediction feature vector output by the prediction model. The prediction model is generated based on the video recommendation model and the target training samples. The video recommendation model is used to recommend videos to users.

[0008] The predicted feature vectors are clustered, and the labels of the video to be processed are determined based on the clustering results.

[0009] In a second aspect of this application, a video tag generation apparatus is provided, the apparatus comprising:

[0010] The first acquisition unit is used to acquire relevant information of the video to be processed, including image, audio and text information of the video to be processed;

[0011] The second acquisition unit is used to input the relevant information into the prediction model to obtain the prediction feature vector output by the prediction model. The prediction model is generated based on the video recommendation model and the target training samples. The video recommendation model is used to recommend videos to users.

[0012] The determining unit is used to cluster the predicted feature vectors and determine the tags of the video to be processed based on the clustering results.

[0013] In a third aspect of this application, an electronic device is provided, the device comprising: a processor and a memory; the memory for storing instructions or computer programs; and the processor for executing the instructions or computer programs in the memory to cause the electronic device to perform the method described in the first aspect.

[0014] In a fourth aspect of this application, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a device, cause the device to perform the method described in the first aspect.

[0015] In a fifth aspect of this application, a computer program product is provided, the computer program product comprising a computer program / instructions that, when executed by a processor, implement the method described in the first aspect.

[0016] Therefore, the embodiments of this application have the following beneficial effects:

[0017] In this embodiment, for any video to be processed, relevant information corresponding to the video is obtained, including image, audio, and text information of the video. This relevant information is input into a prediction model to obtain a predicted feature vector output by the model. The prediction model is generated based on a video recommendation model and target training samples. The video recommendation model is used to recommend videos to users, and during recommendation, it can extract feature vectors reflecting user interests from the videos. That is, the predicted feature vector output by the prediction model can reflect the user's interest features. The predicted feature vectors are clustered, and the tags of the video to be processed are determined based on the clustering results. It is evident that the technical solution provided in this application can obtain the predicted features of any video to be processed, and these predicted features can match the user's interests. Furthermore, clustering based on these predicted features yields a classification tag for the video to be processed. This allows for more accurate tag matching even for videos with less interaction, moving beyond the limitations of expert-defined tag systems and enabling videos to be recommended to more users. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart of a video tag generation method provided in this application embodiment;

[0020] Figure 2 This is a schematic diagram of a video tag generation scenario provided in an embodiment of this application;

[0021] Figure 3 A schematic diagram of a video tag generation device provided in an embodiment of this application;

[0022] Figure 4 This is a schematic diagram of an electronic device structure provided in an embodiment of this application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0024] Since video tags are often generated based on tag systems established by expert definition and operational mining, on the one hand, the granularity of classification is often fixed and it is difficult to adapt to all scenarios; on the other hand, since human interests are fine-grained and complex, it is difficult to achieve accurate classification by tag systems that rely on expert definition.

[0025] Based on this, this application provides a video tag generation method. For videos to be processed that have not yet been assigned tags, relevant information about the video is obtained, including image, audio, and text information, i.e., multimodal information of the video. This relevant information is input into a prediction model, which obtains a predicted feature vector that reflects user interests. The prediction model is trained based on a video recommendation model and target training samples. The video recommendation model is used to recommend videos to users, and it can extract feature vectors reflecting user interests from the videos during recommendation, thus ensuring that the predicted feature vectors output by the trained prediction model also reflect user interests. Clustering is performed based on the predicted feature vectors to obtain clustering results, which are then used to determine the tags of the videos to be processed. In other words, the tags of the videos to be processed are determined by clustering predicted features, eliminating reliance on expert-defined tag systems and improving classification accuracy.

[0026] It is understood that before using the technical solutions of the various embodiments in this disclosure, users will be informed of the type, scope of use, and usage scenarios of the personal information involved in an appropriate manner, and user authorization will be obtained.

[0027] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose, based on the prompt message, whether to provide personal information to the software or hardware such as electronic devices, applications, servers, or storage media performing the operations of this disclosed technical solution.

[0028] As an optional but not limited implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0029] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0030] To facilitate understanding of the technical solutions provided in the embodiments of this application, the following description will be provided in conjunction with the accompanying drawings.

[0031] See Figure 1 This figure illustrates a video tag generation method provided in an embodiment of this application. This method can be executed by a video tag generator, which can be an electronic device or a server. The electronic device can include devices with communication functions such as mobile phones, tablets, laptops, in-vehicle terminals, wearable electronic devices, all-in-one machines, and smart home devices, or it can be a device simulated by a virtual machine or simulator. Figure 1 As shown, the method may include the following steps:

[0032] S101: Obtain relevant information about the video to be processed.

[0033] In this embodiment, to accurately obtain the features of the video to be processed, relevant information is extracted from the video. This relevant information includes image, audio, and text information from the video. That is, the multimodal content of the video to be processed is obtained. The text information may include the title of the video and characters from the video frame, which can be obtained using optical character recognition (OCR) technology.

[0034] S102: Input the relevant information into the prediction model to obtain the prediction feature vector output by the prediction model.

[0035] After extracting relevant information from the video to be processed, this information is input into a prediction model to obtain a predicted feature vector. The prediction model is used to predict feature vectors based on the input relevant information, and the predicted feature vectors can reflect the user's interests.

[0036] The prediction model is generated based on the video recommendation model and target training samples. The video recommendation model is used to recommend videos to users. When recommending videos to users, the feature vectors extracted by the video recommendation model can reflect the user's interests. Therefore, the feature vectors predicted by the prediction model generated based on the video recommendation model can also express the user's interests.

[0037] The training process of the prediction model includes:

[0038] 1) Input the target training samples into the video recommendation model to obtain the recommendation feature vector.

[0039] The video recommendation model is a pre-trained model used to recommend videos of interest to users. To improve training accuracy, the target training samples can be videos frequently accessed by users; that is, the frequency of access to the target training samples is greater than or equal to a preset access frequency threshold. Since frequently accessed data often undergoes more user interactions, its corresponding recommendation feature vectors are more expressive and purer. The frequency of user access to the target training samples meeting the preset condition indicates that the target training samples belong to frequently accessed data. The specific value of the preset access frequency threshold can be determined according to the actual application situation, and is not limited in this embodiment.

[0040] 2) Input the target training samples into the initial prediction model to obtain the initial prediction feature vector.

[0041] 3) Train the initial prediction model based on the recommended feature vector and the initial prediction feature vector until the distance between the initial prediction feature vector and the recommended feature vector is less than a preset distance threshold, and then generate the prediction model.

[0042] At the start of training, the target training samples are input into the initial prediction model to obtain the initial predicted feature vector output by the model. The recommended feature vector is used as the true value of the target training sample, and the predicted feature vector is used as the predicted value of the target training sample to train the initial prediction model, updating the relevant parameters and obtaining an updated initial prediction model. The target training samples are input into the updated initial prediction model again to obtain an updated initial predicted feature vector. The updated initial prediction model is then trained again using the recommended feature vector and the updated initial predicted feature vector, and this training process is repeated until the distance between the initial predicted feature vector output by the initial prediction model and the recommended feature vector is less than a preset distance threshold, at which point the prediction model is obtained.

[0043] In one embodiment of this disclosure, training an initial prediction model based on recommended feature vectors and initial predicted feature vectors includes: training the initial prediction model based on the recommended feature vectors, initial predicted feature vectors, and a loss function until the loss function meets a preset condition, thereby generating a prediction model. That is, the recommended feature vectors and initial predicted feature vectors are input into the loss function, and the initial prediction model is trained based on the value of the loss function until the value of the loss function meets a preset condition. The loss function can be a cosine loss function, a mean squared error loss function, etc.

[0044] S103: Cluster the predicted feature vectors and determine the labels of the videos to be processed based on the clustering results.

[0045] After obtaining the predicted feature vector, the predicted feature vector is clustered to obtain the clustering results, and then the labels of the videos to be processed are determined based on the clustering results. Specifically, the predicted feature vector can be input into a pre-trained clustering model, the clustering results can be obtained through the clustering model, and the labels corresponding to the clustering results can be determined as the labels of the videos to be processed.

[0046] In one embodiment of this disclosure, to improve the diversity of clustering and thus determine diverse classification labels for the video to be processed, this embodiment obtains multiple clustering granularities to determine cluster centers similar to the predicted feature vector at different clustering granularities. Specifically: for any clustering granularity of k types, the distance between the predicted feature vector and each cluster center vector in the clustering granularity is determined, where k is greater than or equal to 2, and the number of classifications corresponding to different clustering granularities is different; based on the distance, m target cluster center vectors are determined from the clustering granularity, where the distance between the target cluster center vector and the predicted feature vector is less than a preset distance threshold, where m is greater than or equal to 1; the labels corresponding to the target cluster center vectors are determined as the labels of the video to be processed. The video to be processed corresponds to k*m labels. That is, the predicted feature vector is compared with different cluster center vectors at different clustering granularities to determine one or more target cluster center vectors from any clustering granularity, and then the labels corresponding to the target cluster center vectors are determined as the labels of the video to be processed. Different clustering granularities correspond to different numbers of categories, different cluster centers correspond to different cluster center vectors, and different cluster center vectors correspond to different labels. For example, a clustering granularity of 64 means that there are 64 categories in this clustering granularity, and a clustering granularity of 512 means that there are 512 categories in this clustering granularity.

[0047] For example, given three clustering granularities of 64, 512, and 2048, for each granularity, three target cluster center vectors are determined based on the predicted feature vector and each cluster center vector within that granularity. This results in a total of nine target cluster center vectors. The labels corresponding to these nine target cluster center vectors are then used as the labels for the video to be processed. In other words, the video to be processed has a total of nine labels.

[0048] In one embodiment of this disclosure, once the tag corresponding to the video to be processed is determined, the video to be processed can be recommended to the user based on the tag, thereby ensuring that the video to be processed can be viewed by interested users.

[0049] In one embodiment of this disclosure, the video to be processed can be not only a video with a high access frequency, but also a video whose access frequency by the user is less than a preset access frequency threshold. That is, the prediction model can perform feature prediction for various types of videos, and make the predicted feature vectors reflect the user's interests. In other words, the prediction model can not only realize the expression of interest in high-popularity content, but also has effective representation capabilities for low-popularity content, thereby dynamically obtaining video tags.

[0050] As can be seen, for any video to be processed, relevant information corresponding to that video is obtained, including image, audio, and text information of the video. This relevant information is input into a prediction model to obtain the predicted feature vector output by the prediction model. The prediction model is generated based on a video recommendation model and target training samples. The video recommendation model is used to recommend videos to users, and during recommendation, it can extract feature vectors reflecting user interests from the videos. That is, the predicted feature vector output by the prediction model can reflect the user's interest features. The predicted feature vectors are clustered, and the tags of the video to be processed are determined based on the clustering results. Therefore, the technical solution provided in this application can obtain the predicted features of any video to be processed, and these predicted features can match the user's interests. Furthermore, clustering based on these predicted features yields the classification tags for the video to be processed. This allows for more accurate tag matching even for videos with less interaction, no longer limited to expert-defined tag systems, thus enabling videos to be recommended to more users.

[0051] For a better understanding of this application, see [link to relevant documentation]. Figure 2 The video tag generation scenario diagram shown first illustrates the process: First, target training samples are input into a video recommendation model and an initial prediction model to obtain a recommendation vector x1 and an initial predicted feature vector x2. Based on x1 and x2, a corresponding loss function (loss) is determined, and the initial prediction model is trained using this loss function. This process is repeated until the loss function meets preset conditions, generating the prediction model. Next, relevant information about the video to be processed is input into the prediction model to obtain a predicted feature vector x. This predicted feature vector x is then input into a clustering model to obtain clustering results c at different clustering granularities. Finally, the tags for the video to be processed are determined based on the clustering results c.

[0052] Based on the above method embodiments, this application provides a video tag generation apparatus and device, which will be described below with reference to the accompanying drawings.

[0053] See Figure 3 This figure is a structural diagram of a video tag generation device provided in an embodiment of this application, such as... Figure 3 As shown, the device includes: a first acquisition unit 301, a second acquisition unit 302, and a determination unit 303.

[0054] The first acquisition unit 301 is used to acquire relevant information of the video to be processed, the relevant information including image, audio and text information of the video to be processed;

[0055] The second acquisition unit 302 is used to input the relevant information into the prediction model to obtain the prediction feature vector output by the prediction model. The prediction model is generated based on the video recommendation model and the target training samples. The video recommendation model is used to recommend videos to users.

[0056] The determining unit 303 is used to cluster the predicted feature vector and determine the label of the video to be processed based on the clustering result.

[0057] In one embodiment of this disclosure, the determining unit 303 is configured to, for any of the k clustering granularities, determine the distance between the predicted feature vector and each cluster center vector in the clustering granularity, wherein k is greater than or equal to 2, and different clustering granularities correspond to different numbers of classifications; determine m target cluster center vectors from the clustering granularity based on the distance, wherein the distance between the target cluster center vectors and the predicted feature vector is less than a preset distance threshold, wherein m is greater than or equal to 1; and determine the labels corresponding to the target cluster center vectors as the labels of the video to be processed, wherein the video to be processed corresponds to k*m labels.

[0058] In one embodiment of this disclosure, the training process of the prediction model includes: inputting the target training sample into the video recommendation model to obtain a recommendation feature vector; inputting the target training sample into an initial prediction model to obtain an initial prediction feature vector; training the initial prediction model based on the recommendation feature vector and the initial prediction feature vector until the distance between the initial prediction feature vector and the recommendation feature vector is less than a preset distance threshold, thereby generating the prediction model.

[0059] In one embodiment of this disclosure, training the initial prediction model based on the recommended feature vector and the initial prediction feature vector includes: training the initial prediction model based on the recommended feature vector, the initial prediction feature vector, and a loss function until the loss function meets a preset condition, thereby generating the prediction model.

[0060] In one embodiment of this disclosure, the target training sample is accessed by the user at a frequency greater than or equal to a preset access frequency threshold.

[0061] In one embodiment of this disclosure, the video to be processed is a video accessed by a user at a frequency less than the preset access frequency threshold.

[0062] In one embodiment of this disclosure, the apparatus further includes: a recommendation unit;

[0063] The recommendation unit is used to recommend the video to be processed to the user based on the tags of the video to be processed.

[0064] It should be noted that the specific implementation of each unit in this embodiment can be found in the relevant descriptions in the above method embodiments.

[0065] The unit division in this embodiment is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used. The functional units in this embodiment can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. For example, in the above embodiment, the processing unit and the sending unit can be the same unit or different units. The integrated unit can be implemented in hardware or as a software functional unit.

[0066] See Figure 4 This diagram illustrates a structural schematic of an electronic device 400 suitable for implementing embodiments of the present disclosure. The terminal devices in these embodiments may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0067] like Figure 4 As shown, electronic device 400 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 401, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 402 or a program loaded from storage device 408 into random access memory (RAM) 404. RAM 403 also stores various programs and data required for the operation of electronic device 400. The processing device 401, ROM 402, and RAM 403 are interconnected via bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0068] Typically, the following devices can be connected to I / O interface 405: input devices 406 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 407 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 408 including, for example, magnetic tapes, hard disks, etc.; and communication devices 409. Communication device 409 allows electronic device 400 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4An electronic device 400 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0069] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 409, or installed from a storage device 408, or installed from a ROM 402. When the computer program is executed by the processing device 401, it performs the functions defined in the methods of embodiments of this disclosure.

[0070] The electronic device provided in this embodiment belongs to the same inventive concept as the method provided in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0071] This disclosure provides a computer storage medium storing a computer program that, when executed by a processor, implements the methods provided in the above embodiments.

[0072] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0073] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0074] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0075] The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, enable the electronic device to perform the aforementioned methods.

[0076] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0077] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0078] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units / modules do not necessarily limit the specific unit itself.

[0079] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0080] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0081] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems or apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

[0082] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0083] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0084] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0085] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method of video tagging, the method comprising: The method comprises: obtaining related information of a to-be-processed video, the related information comprising image, audio and text information of the to-be-processed video; inputting the related information into a prediction model to obtain a prediction feature vector output by the prediction model, the prediction feature vector comprising a feature vector reflecting user interest, the prediction model being generated by training based on a video recommendation model and a target training sample, the video recommendation model being used to recommend videos of interest to a user to the user; clustering the prediction feature vector to determine a label of the to-be-processed video according to a clustering result; wherein the prediction model is obtained after training an initial prediction model based on a recommendation feature vector and an initial prediction feature vector until a distance between the initial prediction feature vector and the recommendation feature vector is less than a preset distance threshold, the recommendation feature vector being obtained by the video recommendation model based on the target training sample, and the initial prediction feature vector being obtained by the initial prediction model based on the target training sample.

2. The method of claim 1, wherein, The clustering of the prediction feature vector and the determination of the label of the to-be-processed video according to the clustering result comprise: for any one of k clustering granularities, determining distances between the prediction feature vector and each clustering center vector in the clustering granularity, wherein k is greater than or equal to 2, and the number of classifications corresponding to different clustering granularities is different; determining m target clustering center vectors from the clustering granularities according to the distances, the distance between the target clustering center vector and the prediction feature vector being less than a preset distance threshold, wherein m is greater than or equal to 1; determining a label corresponding to the target clustering center vector as the label of the to-be-processed video, the to-be-processed video corresponding to k*m labels.

3. The method of claim 1, wherein, The training of the initial prediction model based on the recommendation feature vector and the initial prediction feature vector comprises: training the initial prediction model based on the recommendation feature vector, the initial prediction feature vector and a loss function until the loss function meets a preset condition to generate the prediction model.

4. The method of claim 1, wherein, The frequency at which the target training sample is accessed by a user is greater than or equal to a preset access frequency threshold.

5. The method of claim 4, wherein, The to-be-processed video is a video that is accessed by a user less frequently than the preset access frequency threshold.

6. The method of claim 1, wherein, The method further comprises: recommending the to-be-processed video to a user according to the label of the to-be-processed video.

7. A video tagging apparatus, comprising: The apparatus comprises: a first obtaining unit configured to obtain related information of a to-be-processed video, the related information comprising image, audio and text information of the to-be-processed video; a second obtaining unit configured to input the related information into a prediction model to obtain a prediction feature vector output by the prediction model, the prediction feature vector comprising a feature vector reflecting user interest, the prediction model being generated by training based on a video recommendation model and a target training sample, the video recommendation model being used to recommend videos of interest to a user to the user; a determining unit configured to cluster the prediction feature vector to determine a label of the to-be-processed video according to a clustering result. The prediction model is obtained by training an initial prediction model based on a recommended feature vector and an initial prediction feature vector until a distance between the initial prediction feature vector and the recommended feature vector is less than a preset distance threshold, the recommended feature vector is obtained by the video recommendation model based on the target training sample, and the initial prediction feature vector is obtained by the initial prediction model based on the target training sample.

8. An electronic device, comprising: The device comprises a processor and a memory; The memory is configured to store instructions or computer programs. The processor is configured to execute the instructions or computer programs in the memory, so that the electronic device executes the method in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores instructions, and when the instructions run on the device, the device executes the method in any one of claims 1-6.

Images