CN115757756A

CN115757756A - Content retrieval method, device, medium and electronic equipment

Info

Publication number: CN115757756A
Application number: CN202211528260.7A
Authority: CN
Inventors: 余俊; 周文
Original assignee: Beijing Youzhuju Network Technology Co Ltd
Current assignee: Beijing Youzhuju Network Technology Co Ltd
Priority date: 2022-11-30
Filing date: 2022-11-30
Publication date: 2023-03-07

Abstract

The present disclosure relates to a content retrieval method, apparatus, medium, and electronic device, the method comprising: acquiring a target video; extracting video text information from the information of the target video in multiple modes, and determining a target theme for describing the target video according to the video text information; according to the target theme, target knowledge content related to the target theme is retrieved in a pre-constructed knowledge graph, wherein the target knowledge content is additionally recommended when a client displays the target video, the knowledge graph comprises a plurality of triples, the triples comprise theme nodes, knowledge content nodes and edges for describing the relationship among the nodes, the knowledge content comprises at least one of knowledge content corresponding to a literature genre, knowledge content corresponding to a movie genre and knowledge content corresponding to a building genre, and the accuracy of the retrieved target knowledge content is improved.

Description

Content retrieval method, device, medium and electronic equipment

Technical Field

The present disclosure relates to the field of electronic information technologies, and in particular, to a content retrieval method, apparatus, medium, and electronic device.

Background

The knowledge is in accordance with the civilization direction and is the sum of results of human exploration on the physical world and the mental world, and comprises descriptions of facts and information or skills acquired in education and practice, and the content corresponding to the knowledge can be spread or recommended to the user so that the user can know the knowledge, and the content corresponding to the knowledge is referred to as the knowledge content for short. Knowledge content cannot be independently distributed in a recommendation scene due to the fact that audience groups are small and limited by types of self genres, and needs to be led out in a card or anchor point mode by means of content of other genres, so that recommendation of the knowledge content is achieved.

Therefore, it is crucial how to recommend relevant knowledge content to the user by means of the content of other genres.

Disclosure of Invention

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

In a first aspect, the present disclosure provides a content retrieval method, including:

acquiring a target video;

extracting video text information from the information of the target video in multiple modes, and determining a target theme for describing the target video according to the video text information;

and retrieving target knowledge content related to the target topic from a pre-constructed knowledge graph according to the target topic, wherein the target knowledge content is additionally recommended when a client displays the target video, the knowledge graph comprises a plurality of triples, the triples comprise topic nodes, knowledge content nodes and edges for describing the relationship among the nodes, and the knowledge content comprises at least one of knowledge content corresponding to a literature genre, knowledge content corresponding to a movie genre and knowledge content corresponding to a building genre.

In a second aspect, the present disclosure provides a content retrieval apparatus comprising:

the acquisition module is used for acquiring a target video;

the determining module is used for extracting video text information from the information of the target video in multiple modes and determining a target theme for describing the target video according to the video text information;

the retrieval module is used for retrieving target knowledge content related to the target topic from a pre-constructed knowledge graph according to the target topic, wherein the target knowledge content is additionally recommended when a client displays the target video, the knowledge graph comprises a plurality of triples, each triplet comprises a topic node, a knowledge content node and an edge for describing the relationship among the nodes, and the knowledge content comprises at least one of knowledge content corresponding to a literature genre, knowledge content corresponding to a movie genre and knowledge content corresponding to a building genre.

In a third aspect, the present disclosure provides a computer readable medium having stored thereon a computer program which, when executed by a processing apparatus, performs the steps of the method of the first aspect.

In a fourth aspect, the present disclosure provides an electronic device comprising:

a storage device having a computer program stored thereon;

processing means for executing the computer program in the storage means to carry out the steps of the method of the first aspect.

According to the technical scheme, the video text information is extracted through the information of various modes, the target theme for describing the target video is determined based on the video text information, the accuracy of identifying the video theme is improved, and therefore the accuracy of retrieving related target knowledge content based on the video theme can be improved; in addition, through the knowledge graph, nodes corresponding to topics with the same name but expressing different themes can be constructed, so that the problem that the retrieved knowledge content is inconsistent with the theme of the target video due to text matching is solved, and the accuracy of the retrieved target knowledge content is further improved.

Additional features and advantages of the present disclosure will be set forth in the detailed description which follows.

Drawings

The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and components are not necessarily drawn to scale. In the drawings:

FIG. 1 is a flow diagram illustrating a method of content retrieval according to an exemplary embodiment.

FIG. 2 is a schematic diagram illustrating a knowledge-graph, according to an exemplary embodiment.

FIG. 3 is a schematic diagram illustrating a determination of a target topic in accordance with one illustrative embodiment.

Fig. 4 is a block diagram illustrating a content retrieval device according to an example embodiment.

Fig. 5 is a schematic diagram of a structure of an electronic device according to an exemplary embodiment.

Detailed Description

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for a more complete and thorough understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.

It should be understood that the various steps recited in the method embodiments of the present disclosure may be performed in a different order, and/or performed in parallel. Moreover, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this respect.

The term "include" and variations thereof as used herein are open-ended, i.e., "including but not limited to". The term "based on" is "based, at least in part, on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions for other terms will be given in the following description.

It should be noted that the terms "first", "second", and the like in the present disclosure are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.

It is noted that references to "a", "an", and "the" modifications in this disclosure are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that "one or more" may be used unless the context clearly dictates otherwise.

The names of messages or information exchanged between devices in the embodiments of the present disclosure are for illustrative purposes only, and are not intended to limit the scope of the messages or information.

It is understood that, before the technical solutions disclosed in the embodiments of the present disclosure are used, the user should be informed of the type, the use range, the use scene, etc. of the personal information related to the present disclosure in a proper manner according to the relevant laws and regulations and obtain the authorization of the user.

For example, in response to receiving an active request from a user, a prompt message is sent to the user to explicitly prompt the user that the requested operation to be performed would require the acquisition and use of personal information to the user. Thus, the user can autonomously select whether to provide personal information to software or hardware such as an electronic device, an application program, a server, or a storage medium that performs the operations of the disclosed technical solution, according to the prompt information.

As an alternative but non-limiting implementation manner, in response to receiving an active request from the user, the manner of sending the prompt information to the user may be, for example, a pop-up window manner, and the prompt information may be presented in a text manner in the pop-up window. In addition, a selection control for providing personal information to the electronic device by the user's selection of "agreeing" or "disagreeing" can be carried in the pop-up window.

It is understood that the above notification and user authorization process is only illustrative and not limiting, and other ways of satisfying relevant laws and regulations may be applied to the implementation of the present disclosure.

At the same time, it is understood that the data involved in the present disclosure (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of the relevant laws and regulations and related regulations.

As the background art shows, because the audience population is small and limited by the genre of the subject, the knowledge content cannot be separately distributed in the recommended scene, and needs to be introduced in the form of a card or an anchor point by the content of other genres, so as to realize recommendation of the knowledge content.

In the related art, texts appearing in the main sanctioned contents are usually identified as the subjects of the main sanctioned contents, but the subjects obtained in such a way cannot accurately summarize the subjects of the main sanctioned contents, so that accurate knowledge contents cannot be matched based on inaccurate subjects, and the recommendation of the knowledge contents is inaccurate; after the theme is determined, the knowledge content corresponding to the text corresponding to the theme is retrieved for recommendation by text matching, and the knowledge content retrieved by text matching may not be related to the subject matter represented by the main sanctioned content, that is, the substantial content of the main sanctioned content is not related to the substantial content corresponding to the knowledge content, for example, the subject word is "materia Medica", and the main sanctioned content represents a shuttlecock operation with the materia Medica as the subject song.

In view of this, the embodiment of the present disclosure discloses a content retrieval method, device, medium, and electronic device, which improve the accuracy of recommendation.

Fig. 1 is a flowchart illustrating a content retrieval method according to an exemplary embodiment, which may be applied to an electronic device, such as a mobile terminal, e.g., a mobile phone, a tablet, and a fixed terminal, e.g., a server, a desktop computer, and the like. Referring to fig. 1, the method includes the following steps:

and step S101, acquiring a target video.

It is to be noted that the target video is the above-mentioned subject matter content, that is, the content that can be distributed individually in the recommended scene.

For example, the target video may be understood as a short video or a medium video, and the short video or the medium video may be a video with a video duration less than a certain duration.

For example, the target video may be a video uploaded by the user for sharing.

Step S102, video text information is extracted from the information of the target video in various modes, and a target theme for describing the target video is determined according to the video text information.

The video text information is text information related to the target video, such as text describing a video title of the target video, subtitle text in the target video, text corresponding to audio in the target video, and the like.

It is worth noting that the target theme of the target video is used to describe the subject matter of the target video.

In some embodiments, the plurality of modalities of the target video may include an audio modality, an image modality, and a text modality. For the audio modality, ASR (Automatic Speech Recognition) may be used to extract video text information from the audio of the target video; for the image modality, video text information can be extracted from the video frames of the target video by using OCR (Optical Character Recognition); for the text modality, video text information may be extracted from a video title of the target video.

Step S103, according to a target theme, target knowledge content related to the target theme is retrieved in a pre-constructed knowledge graph, wherein the target knowledge content is additionally recommended when a target video is displayed on a client side, the knowledge graph comprises a plurality of triples, the triples comprise theme nodes, knowledge content nodes and edges for describing the relationship among the nodes, and the knowledge content comprises at least one of knowledge content corresponding to a literature genre, knowledge content corresponding to a movie genre and knowledge content corresponding to a building genre.

The client may be an electronic terminal used by a user. The electronic equipment retrieves the target knowledge content through the target theme of the target video, so that the target video and the target knowledge content can be recommended to the client side together, and the client side can display corresponding electronic data for the user to read and know by clicking the corresponding link of the target knowledge content.

It is noted that the target knowledge contents are deeper and better contents for describing the target subject, such as books, poems, documentaries, teaching videos corresponding to public classes, web links corresponding to museums, and so on.

It should be noted that each node in the knowledge graph uniquely corresponds to one topic, and the nodes may be characterized by an ID (Identity) in the knowledge graph, so that each node uniquely corresponds to one topic. On this basis, in a knowledge graph, for topics with the same name, it is possible to characterize topics expressing different topics with different IDs. The topic node pairs in the knowledge graph are applied to representing topics, and the knowledge node pairs in the knowledge graph are applied to representing knowledge contents.

Illustratively, the knowledge content corresponding to the literature genre includes books and poems, the knowledge content corresponding to the movie genre includes documentaries and teaching videos corresponding to public classes, the knowledge content corresponding to the building genre includes web links corresponding to museums, and the books, poems, documentaries, teaching videos corresponding to public classes and web links corresponding to museums can be used as ways for users to know knowledge.

Fig. 2 is a schematic diagram of a knowledge-graph according to an exemplary embodiment, in fig. 2, each rectangular box represents a node, edges are constructed between the nodes, the edges are illustrated by curves with arrows in fig. 2, the edges between the nodes are used for describing relationships between the nodes, and two nodes and the edges between the two nodes form a triplet. The knowledge content 1 may be a book, the knowledge content 2 may be a documentary, the knowledge content 1 and the knowledge content 2 are knowledge contents corresponding to the topic 1, and the topic 2 may be an author of the knowledge content 1. Taking the knowledge graph shown in fig. 2 as an example, in a target topic of a target video as a topic 1, according to the relationship of edges, knowledge content 1 and knowledge content 2 related to the target topic can be retrieved correspondingly as target knowledge content; when the target subject of the target video is subject 2, the knowledge content 1 related to the target subject 2 can be retrieved correspondingly as the target knowledge content.

It should be noted that, for short videos and medium videos such as target videos, the target videos may be understood as short contents obtained by reducing a certain detailed content, and target knowledge contents related to the subject of the target videos are recommended based on the short videos and the medium videos, so that a user may enter deeper and better long contents based on the short contents, and the user may conveniently know the deeper and better long contents. Here, the long content is the target knowledge content.

By the mode, the video text information is extracted through the information of multiple modes, the target theme for describing the target video is determined based on the video text information, and the accuracy of identifying the video theme is improved, so that the accuracy of retrieving related target knowledge content based on the video theme can be improved; in addition, through the knowledge graph, nodes corresponding to topics with the same name but expressing different themes can be constructed, so that the problem that the retrieved knowledge content is inconsistent with the theme of the target video due to text matching is solved, and the accuracy of the retrieved target knowledge content is further improved.

In some embodiments, the step S102 may be implemented by: extracting video text information from information of multiple modes of a target video; processing the video text information through a pre-trained theme extraction model, and recalling at least one candidate theme; and determining a target subject with the relevance to the target video reaching a preset condition in the at least one candidate subject.

The plurality of modalities of the target video include an audio modality, an image modality, and a text modality. For explanation of extracting the video text information from the information of each modality, reference may be made to the above related embodiments, which are not described herein again.

Illustratively, the topic extraction model may be a NER (Named Entity Recognition) model. The NER model can be a model structure of a bidirectional LSTM (Long Short-Term Memory network) + CRF (Conditional Random Field), the LSTM has strong Long sequence feature extraction capability, and the CRF is used as a decoding tool and can convert abundant information coded by the LSTM into an NER labeling sequence, so that an identified entity is obtained based on the NER labeling sequence. For the training of the NER model, reference may be made to the related art.

Illustratively, the topic extraction model may be a keyword extraction model. The keyword extraction model can be implemented based on a TF-IDF (term frequency-inverse text frequency index) algorithm, a TextRank algorithm, and the like. The related art can be referred to for training of the keyword extraction model.

It is worth noting that the candidate topics are keywords or entities extracted by the topic extraction model. The keywords or entities herein may characterize people, places, and the like.

Note that the preset condition is a condition related to correlation. For setting the preset condition, reference may be made to the following embodiments, which are not described herein again.

By the mode, entities or keywords in the video text information can be extracted from the information of multiple modalities of the target video to serve as recall candidate topics, recall is performed based on multiple modalities to improve the recall rate of the candidate topics, and then the target topics of the target video are determined based on the correlation between the recalled candidate topics and the target video, so that the problem that topics cannot accurately express video topics due to the fact that the video topics are determined only through single-modality information in the correlation technology is solved, and an accurate data basis is provided for retrieving knowledge contents based on the topics.

In some embodiments, the step of determining a target topic having a relevance to the target video meeting a preset condition in the at least one candidate topic may be implemented by: acquiring characteristic information of multiple modes of a target video; processing the characteristic information and at least one candidate topic through a pre-trained relevance calculation model to obtain a relevance probability value corresponding to each candidate topic; and determining the candidate topic corresponding to the relevance probability value of which the relevance probability value reaches the preset condition as the target topic.

And the relevance probability value is used for representing the relevance degree of the candidate topic and the target video, and the higher the relevance probability value is, the higher the relevance degree of the candidate topic and the target video is.

The correlation calculation model can be a neural network model with a transform structure, the transform structure comprises an encoder and a decoder, the encoder encodes the feature information of multiple modes of the target video to obtain encoding features, and the decoder particularly decodes the encoding to obtain a correlation probability value corresponding to each candidate topic. For example, the correlation calculation model can be trained by the following method: acquiring a training sample, extracting characteristic information of multiple modes of the training sample, inputting the characteristic information of the multiple modes of the training sample into a neural network model with a transform structure, calculating a loss value based on the output of the neural network model and a label of the training sample, and updating parameters of the neural network model by using the loss value, thereby obtaining a trained correlation calculation model.

For example, the feature information of the multiple modalities of the target video may include an image frame sequence of the image modality, the image sequence including at least two frames of images extracted from the target video, and each frame of image in the image sequence being ordered in a time sequence; the feature information of the multiple modalities of the target video may include a title feature of a text modality, wherein the title feature is extracted from a video title in the target video; the characteristic information of the multiple modalities of the target video can comprise text characteristics extracted from an image of an image modality, wherein the text characteristics are extracted from characters in the image by using an OCR technology, and the characters can be subtitles; the feature information for the multiple modalities of the target video may include text features extracted from audio of the audio modalities, the text features extracted from the audio using ASR techniques.

It is understood that the feature information of the multiple modalities of the target video input to the correlation calculation model may be one or more of the feature information of each modality in the above examples, and this embodiment is not described herein again.

FIG. 3 is a diagram illustrating a determination of a target topic, according to an example embodiment. Referring to fig. 3, first, recalling a candidate topic corresponding to a target video, for example, inputting a first text, a second text, and a third text into an NER model for entity recognition, and determining the candidate topic based on a result of the entity recognition; then, calculating a relevance probability value of the recalled candidate topics, ranking the candidate topics based on the relevance probability value, and determining a target topic of the target video based on a ranking result and a preset condition, for example, inputting the candidate topics, an image sequence, a first text, a second text and a third text into a relevance calculation model to obtain a relevance probability value corresponding to each candidate topic output by the relevance calculation model; finally, the target topic of the target video is determined based on the relevance probability values of all candidate topics.

Wherein, the first text may be video text information extracted from a video title of the target video; the second text may be video text information extracted from audio of the target video using ASR techniques; the third text may be video text information extracted from video frames of the target video using OCR techniques.

In this embodiment, for the NER model, the correlation calculation model, etc., reference may be made to the above-mentioned related embodiments, which are not described herein again.

In some embodiments, the target topic may be determined according to the relevance probability value and the preset condition by: determining candidate correlation probability values of all correlation probability values located at the front preset bits; and determining the candidate topic corresponding to the candidate relevance probability value with the candidate relevance probability value exceeding the first preset probability threshold value as the target topic.

For example, all the correlation probability values may be sorted in an order from high to low to obtain a sorting result; and taking the candidate relevance probability value of the front preset bit in the sequencing result, and determining the candidate theme corresponding to the candidate relevance probability value exceeding a first preset probability threshold in the candidate relevance probability value of the front preset bit as the target theme.

The first preset probability threshold may be set through actual conditions, which is not limited in this embodiment.

Through the method, the first probability threshold value used for filtering the candidate theme with low correlation is set, so that the aim that the target theme can well express the target video can be guaranteed, and the accuracy of identifying the video theme is improved; and the candidate relevance probability value of the front preset bit is taken out by using a first sorting mode, and then the candidate relevance probability value is compared with the first probability threshold value in terms of magnitude relation, so that the problem of high resource occupation caused by comparing the relevance probability value of the candidate topics with the first preset probability threshold value one by one when the number of the candidate topics is large is solved.

In some embodiments, the target topic may be determined according to the relevance probability value and the preset condition by: and determining the candidate topic corresponding to the relevance probability value with the relevance probability value exceeding a second preset probability threshold value as the target topic.

The second preset probability threshold may be set according to actual conditions, which is not limited in this embodiment. The first preset probability threshold and the second preset probability threshold may be the same or different.

Through the mode, the second probability threshold value used for filtering the candidate theme with low correlation is set, so that the aim that the target theme can well express the target video can be guaranteed, and the accuracy of identifying the video theme is improved.

Fig. 4 is a block diagram illustrating a content retrieval apparatus according to an exemplary embodiment, and referring to fig. 4, the content retrieval apparatus 400 includes:

an obtaining module 401, configured to obtain a target video;

a determining module 402, configured to extract video text information from information of multiple modalities of the target video, and determine a target topic for describing the target video according to the video text information;

a retrieving module 403, configured to retrieve, according to the target topic, a target knowledge content related to the target topic from a pre-constructed knowledge graph, where the target knowledge content is additionally recommended when the target video is displayed on a client, the knowledge graph includes multiple triples, each triplet includes a topic node, a knowledge content node, and an edge for describing a relationship between the nodes, and the knowledge content includes at least one of a knowledge content corresponding to a genre of literature, a knowledge content corresponding to a genre of movie, and a knowledge content corresponding to a genre of building.

Optionally, the determining module 402 includes:

the extraction submodule is used for extracting video text information from information of multiple modals of the target video, wherein the multiple modals comprise an audio modality, an image modality and a text modality;

the recall submodule is used for processing the video text information through a pre-trained theme extraction model and recalling at least one candidate theme;

and the determining sub-module is used for determining a target topic of which the correlation with the target video reaches a preset condition in the at least one candidate topic.

Optionally, the determining submodule is configured to: acquiring characteristic information of multiple modes of the target video; processing the feature information and the at least one candidate topic through a pre-trained relevance calculation model to obtain a relevance probability value corresponding to each candidate topic; and determining the candidate topic corresponding to the relevance probability value of which the relevance probability value reaches the preset condition as the target topic.

Optionally, the topic extraction model is a named entity recognition model or a keyword extraction model.

Optionally, the determining sub-module is further configured to: determining candidate relevance probability values of all the relevance probability values located at the front preset positions; and determining the candidate topic corresponding to the candidate relevance probability value with the candidate relevance probability value exceeding a first preset probability threshold value as the target topic.

Optionally, the determining sub-module is further configured to: and determining the candidate topic corresponding to the relevance probability value of which the relevance probability value exceeds a second preset probability threshold value as a target topic.

Optionally, the knowledge content that literature genre corresponds includes books and poetry, the knowledge content that movie genre corresponds includes documentary and the teaching video that the public class corresponds, the knowledge content that building genre corresponds includes the corresponding webpage link of museum.

The embodiments of the modules in the apparatus may refer to the related embodiments, which are not described herein again.

The disclosed embodiments also provide a computer readable medium having stored thereon a computer program that, when executed by a processing device, performs the steps of the above-described content retrieval method.

An embodiment of the present disclosure further provides an electronic device, including:

a storage device having a computer program stored thereon;

processing means for executing the computer program in the storage means to implement the steps of the above-mentioned content retrieval method.

Referring now to FIG. 5, a block diagram of an electronic device 500 suitable for use in implementing embodiments of the present disclosure is shown. The terminal device in the embodiments of the present disclosure may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet computer), a PMP (portable multimedia player), a vehicle terminal (e.g., a car navigation terminal), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. The electronic device shown in fig. 5 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.

As shown in fig. 5, electronic device 500 may include a processing means (e.g., central processing unit, graphics processor, etc.) 501 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 502 or a program loaded from a storage means 508 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data necessary for the operation of the electronic apparatus 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.

Generally, the following devices may be connected to the I/O interface 505: input devices 506 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 507 including, for example, a Liquid Crystal Display (LCD), speakers, vibrators, and the like; storage devices 508 including, for example, magnetic tape, hard disk, etc.; and a communication device 509. The communication means 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. While fig. 5 illustrates an electronic device 500 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided.

In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present 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 method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication means 509, or installed from the storage means 508, or installed from the ROM 502. The computer program performs the above-described functions defined in the methods of the embodiments of the present disclosure when executed by the processing device 501.

It should be noted that the computer readable medium of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.

In some implementations, the electronic devices may communicate using any currently known or future developed network Protocol, such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.

The computer readable medium may be embodied in the electronic device; or may be separate and not incorporated into the electronic device.

The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: acquiring a target video; extracting video text information from the information of the target video in multiple modes, and determining a target theme for describing the target video according to the video text information; and retrieving target knowledge content related to the target topic from a pre-constructed knowledge graph according to the target topic, wherein the target knowledge content is additionally recommended when a client displays the target video, the knowledge graph comprises a plurality of triples, the triples comprise topic nodes, knowledge content nodes and edges for describing the relationship among the nodes, and the knowledge content comprises at least one of knowledge content corresponding to a literature genre, knowledge content corresponding to a movie genre and knowledge content corresponding to a building genre.

Computer program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages, including but not limited to an object oriented programming language such as Java, smalltalk, C + +, and including conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present disclosure may be implemented by software or hardware. Wherein the name of a module in some cases does not constitute a limitation on the module itself.

The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems on a chip (SOCs), complex Programmable Logic Devices (CPLDs), and the like.

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

The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other embodiments in which any combination of the features described above or their equivalents does not depart from the spirit of the disclosure. For example, the above features and (but not limited to) the features disclosed in this disclosure having similar functions are replaced with each other to form the technical solution.

Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.

Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims. With regard to the apparatus in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be described in detail here.

Claims

1. A method for retrieving content, comprising:

acquiring a target video;

2. The method according to claim 1, wherein the extracting video text information from the multi-modal information of the target video and determining a target subject describing the target video according to the video text information comprises:

extracting video text information from information of a plurality of modalities of the target video, wherein the plurality of modalities comprise an audio modality, an image modality and a text modality;

processing the video text information through a pre-trained theme extraction model, and recalling at least one candidate theme;

and determining a target subject of which the correlation with the target video reaches a preset condition in the at least one candidate subject.

3. The method of claim 2, wherein the determining a target topic having a relevance to the target video that meets a preset condition in the at least one candidate topic comprises:

acquiring characteristic information of multiple modes of the target video;

processing the feature information and the at least one candidate topic through a pre-trained relevance calculation model to obtain a relevance probability value corresponding to each candidate topic;

and determining the candidate subject corresponding to the relevance probability value reaching the preset condition as the target subject.

4. The method of claim 2, wherein the topic extraction model is a named entity recognition model or a keyword extraction model.

5. The method of claim 3, wherein determining the candidate topic corresponding to the relevance probability value that reaches the preset condition as the target topic comprises:

determining candidate correlation probability values of all the correlation probability values positioned at the front preset bits;

and determining the candidate topic corresponding to the candidate relevance probability value with the candidate relevance probability value exceeding a first preset probability threshold value as the target topic.

6. The method of claim 3, wherein determining the candidate topic corresponding to the relevance probability value that reaches the preset condition as the target topic comprises:

and determining the candidate topic corresponding to the relevance probability value of which the relevance probability value exceeds a second preset probability threshold value as a target topic.

7. The method according to any one of claims 1 to 6, wherein the knowledge content corresponding to the literature genre comprises books and poems, the knowledge content corresponding to the movie genre comprises documentaries and teaching videos corresponding to public classes, and the knowledge content corresponding to the building genre comprises web page links corresponding to museums.

8. A content retrieval apparatus, comprising:

the acquisition module is used for acquiring a target video;

9. A computer-readable medium, on which a computer program is stored, characterized in that the program, when being executed by processing means, carries out the steps of the method of any one of claims 1 to 7.

10. An electronic device, comprising:

a storage device having a computer program stored thereon;

processing means for executing the computer program in the storage means to carry out the steps of the method according to any one of claims 1 to 7.