WO2021135358A1 - 视频分发时效的确定方法和装置 - Google Patents

视频分发时效的确定方法和装置 Download PDF

Info

Publication number
WO2021135358A1
WO2021135358A1 PCT/CN2020/113970 CN2020113970W WO2021135358A1 WO 2021135358 A1 WO2021135358 A1 WO 2021135358A1 CN 2020113970 W CN2020113970 W CN 2020113970W WO 2021135358 A1 WO2021135358 A1 WO 2021135358A1
Authority
WO
WIPO (PCT)
Prior art keywords
video
period
distributed
timeliness
title
Prior art date
Application number
PCT/CN2020/113970
Other languages
English (en)
French (fr)
Inventor
于天宝
邓天生
杜鹏
贠挺
陈国庆
陈政男
Original Assignee
百度在线网络技术（北京）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 百度在线网络技术（北京）有限公司 filed Critical 百度在线网络技术（北京）有限公司
Priority to KR1020217008845A priority Critical patent/KR102542232B1/ko
Priority to EP20862000.5A priority patent/EP3872651A4/en
Priority to JP2021517017A priority patent/JP7140913B2/ja
Priority to US17/186,270 priority patent/US20210192222A1/en
Publication of WO2021135358A1 publication Critical patent/WO2021135358A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/73—Querying
    • G06F16/735—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/7834—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using audio features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/7844—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using original textual content or text extracted from visual content or transcript of audio data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/7867—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings

Definitions

  • This application relates to video distribution technology in the field of data processing, and in particular to a method and device for determining the timeliness of video distribution.
  • the information flow recommendation system is used to provide users with continuously updated media resources. Among them, video accounts for a high proportion. Unlike active search of videos, the videos recommended by the information flow recommendation system to users should not make users feel out of date. video. In other words, each video should have a time limit suitable for its distribution, and the recommendation system should stop distributing the corresponding video after this time limit period.
  • the timeliness of all videos is manually marked. For example, reviewers mark the timeliness of a video as strong timeliness based on subjective experience, and the recommendation system then distributes the video within the time period corresponding to the strong timeliness.
  • a method and device for determining the timeliness of video distribution are provided.
  • a method for determining the time limit of video distribution which is applied to a device for determining the time limit of video distribution, and the method includes:
  • the aging end time of the video to be distributed According to the initial release time of the video to be distributed and the aging period, determine the aging end time of the video to be distributed.
  • the video-related information includes at least one of the following information: title, author, voice, video, subtitles, characters, and title syntax analysis.
  • the determining the time period of the video to be distributed according to the video-related information includes:
  • the third period is determined by identifying keywords of at least one of the speech, the subtitles, the characters, and the caption syntax analysis;
  • the period with the shortest duration among the first period, the second period, the third period, and the fourth period is determined as the aging period of the video to be distributed.
  • each information in the video-related information may correspond to a valid duration
  • this application provides the above four ways to determine the period. After the corresponding period is determined through these four methods, the four periods are referred to to finally determine the video to be distributed.
  • the aging period solves the problem of low accuracy caused by a single method to determine the final aging period.
  • the determining the first period by text classification of the title includes:
  • the first period is determined according to the text category corresponding to the title and the first mapping relationship, and the first mapping relationship represents the correspondence between the text category and the period.
  • the determining the second period by performing news identification on the video includes:
  • the second period is determined according to the video type corresponding to the video and a second mapping relationship, where the second mapping relationship represents a correspondence between the video type and the period.
  • the video types in the second mapping relationship include: news; news videos have the shortest period.
  • the determining the third period by recognizing keywords of at least one of the speech, the subtitles, the characters, and the caption syntax analysis includes:
  • the period with the shortest duration among the periods corresponding to each of the at least one keyword is determined as the third period.
  • the determination of the fourth cycle according to the author includes:
  • the fourth period is determined according to the author of the to-be-distributed video and the fourth mapping relationship, and the fourth mapping relationship represents the correspondence between the author and the period.
  • the present application provides a video distribution time limit determination device, including:
  • the parsing module is used to analyze the video to be distributed to obtain the video related information of the video to be distributed;
  • a timeliness policy module configured to determine the timeliness period of the video to be distributed according to the video-related information, where the timeliness period is used to indicate the effective duration corresponding to the video to be distributed;
  • the timeliness decision module is configured to determine the timeliness end time of the video to be distributed according to the initial release time of the video to be distributed and the timeliness period.
  • the video-related information includes at least one of the following information: title, author, voice, video, subtitles, characters, and title syntax analysis.
  • the timeliness strategy module includes: a headline text classification unit, a news recognition unit, a keyword recognition unit, an author recognition unit, and a determination unit;
  • the title text classification unit is used to determine the first period by text classification of the title
  • the news identification unit is configured to determine the second cycle by performing news identification on the video
  • the keyword recognition unit is configured to identify the third period by recognizing keywords of at least one of the speech, the subtitles, the characters, and the caption syntax analysis;
  • the author identification unit is used to determine the fourth cycle according to the author.
  • the determining unit is configured to determine the period with the shortest duration among the first period, the second period, the third period, and the fourth period as the aging period of the video to be distributed.
  • title text classification unit is specifically used for:
  • the first period is determined according to the text category corresponding to the title and the first mapping relationship, and the first mapping relationship represents the correspondence between the text category and the period.
  • the news identification unit is specifically configured to:
  • the second period is determined according to the video type corresponding to the video and a second mapping relationship, where the second mapping relationship represents a correspondence between the video type and the period.
  • the video types in the second mapping relationship include: news; news videos have the shortest period.
  • the keyword identification unit is specifically configured to:
  • the period with the shortest duration among the periods corresponding to each of the at least one keyword is determined as the third period.
  • the author identification unit is specifically used for:
  • the fourth period is determined according to the author of the to-be-distributed video and the fourth mapping relationship, and the fourth mapping relationship represents the correspondence between the author and the period.
  • this application provides an electronic device, including:
  • At least one processor At least one processor
  • a memory communicatively connected with the at least one processor; wherein,
  • the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the above video distribution timeliness determination method.
  • the present application provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the above video distribution timeliness determination method.
  • the method and device for determining the timeliness of video distribution provided in this application. It does not rely on the reviewer’s timeliness labeling experience. Therefore, it is not interfered by the reviewers’ subjective factors, the labeling quality is stable, and there is no need to train the reviewers, which saves manpower and material resources and also improves the efficiency of determining the timeliness of the video.
  • FIG. 1 is a schematic flowchart of Embodiment 1 of the method for determining the timeliness of video distribution provided by this application;
  • Embodiment 2 is a schematic flowchart of Embodiment 2 of the method for determining the timeliness of video distribution provided by this application;
  • FIG. 3 is a schematic diagram of the first mapping relationship provided by this application.
  • FIG. 4 is a schematic diagram of the second mapping relationship provided by this application.
  • FIG. 5 is a schematic diagram of the third mapping relationship provided by this application.
  • FIG. 6 is a schematic diagram of the fourth mapping relationship provided by this application.
  • FIG. 7 is a schematic structural diagram of an embodiment of an apparatus for determining the timeliness of video distribution provided by this application.
  • FIG. 8 is a block diagram of the electronic device provided by this application.
  • the videos recommended by the information flow recommendation system to users should be videos that do not make users feel outdated.
  • the timeliness of each video is manually marked, and the recommendation system is further based on The timeliness of manual labeling completes the distribution of videos. For example, reviewers mark the timeliness of a video as strong timeliness based on subjective experience, and the recommendation system then distributes the video within the time period corresponding to the strong timeliness.
  • the manual labeling method will have the following problems: subjective factors are seriously interfered with, different reviewers may have different labeling results for the same video, and the labeling quality is unstable; the flexibility is poor, and the access and labeling standards of new video resources The changes require the retraining of reviewers, which consumes a lot of manpower and material resources; the video is of huge magnitude, and the manual labeling method is very inefficient.
  • this application provides a method and device for determining the timeliness of video distribution.
  • the timeliness determining device can be integrated in the information flow recommendation system. After the video to be distributed is input to the device, the device can Determine the timeliness of the video to be distributed by executing the method for determining the timeliness of video distribution provided in this application, and pass the timeliness to the module for distributing videos of the information flow recommendation system. The module completes the distribution of the video, and the entire process does not require manual labor. Intervention can solve the above-mentioned technical problems in the prior art.
  • the time-determination device After the time-determination device receives the video to be distributed, it parses and obtains the video-related information of the video to be distributed, such as: Title, author, voice, video, subtitles, etc.; since each of these video-related information may correspond to a period, you can determine multiple periods based on these video-related information, and then take the shortest period of time as the video to be distributed Finally, combined with the initial release time of the video to be distributed, the time limit end time of the video to be distributed is determined, and the time limit end time is passed to the module that distributes the video, so that the module completes the above to be distributed before the time limit end time Video distribution.
  • the time-determination device After the time-determination device receives the video to be distributed, it parses and obtains the video-related information of the video to be distributed, such as: Title, author, voice, video, subtitles, etc.; since each of these video-related information may correspond to a period, you can determine multiple periods based on these video-related information, and
  • FIG. 1 is a schematic flowchart of Embodiment 1 of the method for determining the timeliness of video distribution provided by this application.
  • the method for determining the timeliness of video distribution provided in this embodiment can be executed by the timeliness determining device described above.
  • the method for determining the timeliness of video distribution provided in this embodiment includes:
  • the video-related information of the video to be distributed may include at least one of the following information: title, author, voice, video, subtitles, characters, and title syntax analysis.
  • the aging determining device parses the file name of the to-be-distributed video to obtain the title and author of the to-be-distributed video, and parses the to-be-distributed video to obtain the to-be-distributed video.
  • the voice, subtitles and video of the video where the parsed video can be the video itself to be distributed with voice and subtitles, or it can be a cartoon clip without voice and subtitles.
  • NLP Natural Language Processing
  • the neural network is used to perform face recognition on the characters in the video. character.
  • the file format corresponding to the title, author, character, and title syntax analysis is a text format, such as .txt, .doc., .docx, .wps, etc.
  • the file format corresponding to the voice is an audio format, such as: MP3, MIDI, WMA, VQF, AMR, etc.
  • the file format corresponding to the subtitle is a picture format, such as: bmp, jpg, png, tif, gif, etc.
  • the file format corresponding to the video is a video format, such as: wmv, asf, asx, etc.
  • S102 Determine a aging period of the video to be distributed according to the video-related information, where the aging period is used to indicate a valid duration corresponding to the video to be distributed.
  • a corresponding period can be determined based on the title, author, voice, video, subtitle, character, and each information in the syntactic analysis of the title obtained by S101. After multiple periods are obtained, the shortest duration is selected. The period of is used as the aging period of the video to be distributed in S102.
  • the period corresponding to each of the foregoing information can be determined through methods such as text classification, news recognition, keyword recognition, and author recognition.
  • S103 Determine the time limit end time of the video to be distributed according to the initial release time of the video to be distributed and the time limit period.
  • the initial release time in step S103 can be obtained when analyzing the video to be distributed, or after analyzing the title, author, person and other information, extract the search keywords from the information and search in the search engine The keyword query is obtained.
  • the time limit ends with the first release time plus 2 days, that is, October 3, 2019.
  • the video to be distributed is parsed to obtain video-related information of the video to be distributed, and then based on the video-related information, the video to be distributed is determined The aging period. Finally, according to the initial release time of the video to be distributed and the above aging period, the aging end time of the video to be distributed is determined.
  • the above method is executed by the device and does not rely on the timeliness of the reviewer Marking experience, therefore, is not disturbed by the subjective factors of the reviewers, the marking quality is stable, and there is no need to train the reviewers, which saves manpower and material resources and improves the efficiency of determining the timeliness of the video.
  • FIG. 2 is a schematic flowchart of Embodiment 2 of the method for determining the timeliness of video distribution provided by this application. As shown in Figure 2, the method for determining the timeliness of video distribution provided in this embodiment includes:
  • S201 Analyze the video to be distributed to obtain video-related information of the video to be distributed.
  • the above-mentioned video-related information includes at least one of the following information: title, author, voice, video, subtitles, characters, and title syntax analysis.
  • title For the implementation of S201, please refer to the first embodiment, which will not be repeated in this application.
  • S202 Determine the first period by performing text classification on the title.
  • the title is first classified by a fast text classification model to obtain the text category corresponding to the title.
  • the first period is determined according to the text category corresponding to the title and the first mapping relationship, and the first mapping relationship represents the correspondence between the text category and the period.
  • FIG. 3 is a schematic diagram of the first mapping relationship.
  • the text category may include, for example, emotion, military, politics, education, and entertainment.
  • the title is input into the fast text classification model.
  • the first mapping relationship shown in Figure 3 matches entertainment with strong timeliness and strong
  • the aging cycle is 3 days, and the first cycle can be determined to be 3 days.
  • S203 Determine the second period by performing news identification on the above-mentioned video.
  • the video is first identified through a neural network model to obtain the video type corresponding to the video. Then, the second period is determined according to the video type corresponding to the video and the second mapping relationship, and the second mapping relationship represents the corresponding relationship between the video type and the period.
  • FIG. 4 is a schematic diagram of the second mapping relationship.
  • the video types may include, for example, variety shows, news, weather forecasts, and movies.
  • the video is input to the neural network model.
  • the second mapping relationship shown in Figure 4 matches the news category with super timeliness.
  • the super aging cycle is 2 days, and the second cycle can be determined to be 2 days.
  • S204 Determine a third period by identifying keywords of at least one of the voice, the subtitles, the characters, and the caption syntax analysis.
  • the third period can be determined by recognizing the keywords of any one of the speech, subtitles, characters, and title syntax analysis, for example, only the voice keywords are recognized to determine the third period. cycle.
  • the third cycle can also be determined by identifying any two or more keywords in speech, subtitles, characters, and title syntax analysis, such as recognizing all keywords of speech, subtitles, characters, and title syntax analysis. Determine the third cycle. Since different keywords correspond to different periods, when multiple keywords are identified, multiple periods will be obtained, and the shortest period of the multiple periods can be used as the third period.
  • first convert speech into first text information through voice recognition technology convert subtitles into second text information through image text recognition technology, and then analyze at least one of characters, title syntax, first text information, and second text information Then, according to the identified at least one keyword and the third mapping relationship, the period corresponding to the at least one keyword is determined, and the period with the shortest duration is determined as the third period.
  • Fig. 5 is a schematic diagram of the third mapping relationship.
  • the keywords may include, for example, finance, automobile, food, games, tourism, Buffett, and decline.
  • the speech recognition technology is used to convert the speech into the first text information
  • the image text recognition technology is used to convert the subtitles into the second text information.
  • the identified keyword is finance
  • the keyword identified from the second text information is automobile
  • the keyword identified from the character is Buffett
  • the keyword identified from the headline syntax analysis is down.
  • the third mapping relationship it can be determined that the period corresponding to finance and decline is 2 days, and the period corresponding to automobiles and Buffett is 3 days. Then the cycle with the shortest duration can be determined as the third cycle, that is, 2 days.
  • the above example takes the keyword recognition of all items of speech, subtitles, characters, and title syntax analysis as an example to illustrate the determination process of the third cycle. As described above, you can also select one or more of them. To determine the third cycle, this application will not repeat them one by one.
  • the fourth period is determined according to the author of the to-be-distributed video and a fourth mapping relationship, and the fourth mapping relationship represents the corresponding relationship between the author and the period.
  • Figure 6 is a schematic diagram of the first mapping relationship. See Figure 6.
  • Authors can include, for example, author A, author B, author C, and author D.
  • the timelessness involved in Figure 3 to Figure 6 refers to: the video to be distributed has no timeliness requirement, and it can be distributed within any length of time.
  • S206 Determine the period with the shortest duration among the first period, the second period, the third period, and the fourth period as the aging period of the video to be distributed.
  • the first period determined by text classification of the title is 3 days; the second period determined by the news recognition of the video is 2 days; by analyzing at least one of voice, subtitles, characters, and title syntax
  • the third cycle for identifying the keywords of is 2 days; according to the author, the fourth cycle is 2 days.
  • the shortest period is 2 days, and the time period of the video to be distributed can be determined to be 2 days.
  • S207 Determine the time limit end time of the video to be distributed according to the initial release time of the video to be distributed and the time limit period.
  • the video distribution aging determination method provided in this embodiment provides an achievable way to determine the corresponding period based on different video-related information. After determining multiple periods, refer to the multiple periods to finally determine the aging period of the video to be distributed. It solves the problem of low accuracy brought by the single method to determine the final time period.
  • FIG. 7 is a schematic structural diagram of an embodiment of a video distribution timeliness determining apparatus 700 provided by this application. As shown in FIG. 7, the video distribution timeliness determining apparatus 700 provided in this embodiment includes:
  • the parsing module 701 is configured to analyze the video to be distributed to obtain video-related information of the video to be distributed;
  • the timeliness policy module 702 is configured to determine the timeliness period of the video to be distributed according to the video-related information, and the timeliness period is used to indicate the effective duration corresponding to the video to be distributed;
  • the timeliness decision module 703 is configured to determine the timeliness end time of the video to be distributed according to the initial release time of the video to be distributed and the timeliness period.
  • the video-related information includes at least one of the following information: title, author, voice, video, subtitles, characters, and title syntax analysis.
  • the timeliness strategy module 702 includes: a headline text classification unit 7021, a news recognition unit 7022, a keyword recognition unit 7023, an author recognition unit 7024, and a determination unit 7025;
  • the title text classification unit 7021 is configured to determine the first period by text classification of the title
  • the news recognition unit 7022 is configured to determine the second cycle by performing news recognition on the video
  • the keyword recognition unit 7023 is configured to identify the third period by recognizing keywords of at least one of the speech, the subtitles, the characters, and the caption syntax analysis;
  • the author identification unit 7024 is configured to determine the fourth cycle according to the author.
  • the determining unit 7025 is configured to determine the period with the shortest duration among the first period, the second period, the third period, and the fourth period as the aging period of the video to be distributed.
  • title text classification unit 7021 is specifically used to:
  • the first period is determined according to the text category corresponding to the title and the first mapping relationship, and the first mapping relationship represents the correspondence between the text category and the period.
  • the news identification unit 7022 is specifically configured to:
  • the second period is determined according to the video type corresponding to the video and a second mapping relationship, where the second mapping relationship represents a correspondence between the video type and the period.
  • the video types in the second mapping relationship include: news; news videos have the shortest period.
  • the keyword identification unit 7023 is specifically configured to:
  • the period with the shortest duration among the periods corresponding to each of the at least one keyword is determined as the third period.
  • the author identification unit 7024 is specifically used to:
  • the fourth period is determined according to the author of the to-be-distributed video and the fourth mapping relationship, and the fourth mapping relationship represents the correspondence between the author and the period.
  • the video distribution timeliness determining device provided in this embodiment can be used to execute the corresponding steps in any of the foregoing method embodiments, and its implementation principles and technical effects are similar, and will not be repeated here.
  • the present application also provides an electronic device and a readable storage medium.
  • FIG. 8 it is a block diagram of an electronic device according to the method for determining the timeliness of video distribution according to an embodiment of the present application.
  • Electronic devices are intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers.
  • Electronic devices can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices.
  • the components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the application described and/or required herein.
  • the electronic device includes: one or more processors 801, a memory 802, and interfaces for connecting various components, including a high-speed interface and a low-speed interface.
  • the various components are connected to each other using different buses, and can be installed on a common motherboard or installed in other ways as needed.
  • the processor may process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of the GUI on an external input/output device (such as a display device coupled to an interface).
  • an external input/output device such as a display device coupled to an interface.
  • multiple processors and/or multiple buses can be used with multiple memories and multiple memories.
  • multiple electronic devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).
  • a processor 801 is taken as an example.
  • the memory 802 is a non-transitory computer-readable storage medium provided by this application.
  • the memory stores instructions executable by at least one processor, so that the at least one processor executes the method for determining the timeliness of video distribution provided by this application.
  • the non-transitory computer-readable storage medium of the present application stores computer instructions, and the computer instructions are used to make the computer execute the method for determining the timeliness of video distribution provided by the present application.
  • the memory 802 as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions/modules (for example, , The analysis module 701, the time-sensitive strategy module 702, and the time-sensitive decision module 703 shown in FIG. 7).
  • the processor 801 executes various functional applications and data processing of the server by running non-transitory software programs, instructions, and modules stored in the memory 802, that is, implements the video distribution timeliness determination method in the foregoing method embodiment.
  • the memory 802 may include a storage program area and a storage data area.
  • the storage program area may store an operating system and an application program required by at least one function; the storage data area may store usage requirements of the electronic device used to implement the method for determining the timeliness of video distribution. Created data, etc.
  • the memory 802 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices.
  • the memory 802 may optionally include a memory remotely provided with respect to the processor 801, and these remote memories may be connected to an electronic device for implementing the method for determining the timeliness of video distribution through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
  • the electronic device of the method for determining the timeliness of video distribution may further include: an input device 803 and an output device 804.
  • the processor 801, the memory 802, the input device 803, and the output device 804 may be connected through a bus or other methods. In FIG. 8, the connection through a bus is taken as an example.
  • the input device 803 can receive input digital or character information, and generate key signal input related to the user settings and function control of the electronic device used to realize the video distribution timeliness determination method, such as touch screen, keypad, mouse, trackpad, touch Input devices such as pads, pointing sticks, one or more mouse buttons, trackballs, joysticks, etc.
  • the output device 804 may include a display device, an auxiliary lighting device (for example, LED), a tactile feedback device (for example, a vibration motor), and the like.
  • the display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some embodiments, the display device may be a touch screen.
  • Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: being implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, the programmable processor It can be a dedicated or general-purpose programmable processor that can receive data and instructions from the storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device. An output device.
  • machine-readable medium and “computer-readable medium” refer to any computer program product, device, and/or device used to provide machine instructions and/or data to a programmable processor ( For example, magnetic disks, optical disks, memory, programmable logic devices (PLD)), including machine-readable media that receive machine instructions as machine-readable signals.
  • machine-readable signal refers to any signal used to provide machine instructions and/or data to a programmable processor.
  • the systems and techniques described here can be implemented on a computer that has: a display device for displaying information to the user (for example, a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) ); and a keyboard and a pointing device (for example, a mouse or a trackball) through which the user can provide input to the computer.
  • a display device for displaying information to the user
  • LCD liquid crystal display
  • keyboard and a pointing device for example, a mouse or a trackball
  • Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (for example, visual feedback, auditory feedback, or tactile feedback); and can be in any form (including Acoustic input, voice input, or tactile input) to receive input from the user.
  • the systems and technologies described herein can be implemented in a computing system that includes back-end components (for example, as a data server), or a computing system that includes middleware components (for example, an application server), or a computing system that includes front-end components (for example, A user computer with a graphical user interface or a web browser, through which the user can interact with the implementation of the system and technology described herein), or includes such back-end components, middleware components, Or any combination of front-end components in a computing system.
  • the components of the system can be connected to each other through any form or medium of digital data communication (for example, a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), and the Internet.
  • the computer system can include clients and servers.
  • the client and server are generally far away from each other and usually interact through a communication network.
  • the relationship between the client and the server is generated by computer programs that run on the corresponding computers and have a client-server relationship with each other.

Abstract

视频分发时效的确定方法和装置，涉及数据处理领域中视频分发技术。对待分发视频进行解析，得到所述待分发视频的视频相关信息（S101）；根据所述视频相关信息，确定所述待分发视频的时效周期，所述时效周期用于表示所述待分发视频对应的有效时长（S102）；根据所述待分发视频的首发时间和所述时效周期，确定所述待分发视频的时效结束时间（S103）。不依赖审核人员的时效性标注经验，因此，不受审核人员主观因素干扰，标注质量稳定，也无需对审核人员进行培训，节省了人力物力，同时也提升了视频时效的确定效率。

Description

视频分发时效的确定方法和装置
本申请要求于2020年01月03日提交中国专利局、申请号为202010005825.8、申请名称为“视频分发时效的确定方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及数据处理领域中的视频分发技术，尤其涉及一种视频分发时效的确定方法和装置。
背景技术
信息流推荐系统用于给用户提供持续更新的媒体资源，其中，视频占比很高，不同于视频的主动搜索，信息流推荐系统推荐给用户的视频应当是不会让用户觉得有过期感的视频。也就是说，每个视频都应该有适合其分发的时效期，过了这段时效期后推荐系统应停止分发对应视频。
现有技术中，采用人工方式对全量视频的时效性进行标注，比如：审核人员根据主观经验将某视频的时效性标注为强时效，推荐系统进而在该强时效对应的时长内分发该视频，
然而，人工标注的方法受主观因素干扰严重，不同的审核人员对同一视频的标注结果可能是不同的，标注质量不稳定，而且，视频量级巨大，人工标注的方式效率很低。
发明内容
提供了一种视频分发时效的确定方法和装置。
根据第一方面，提供了一种视频分发时效确定方法，应用于视频分发时效确定装置，该方法包括：
对待分发视频进行解析，得到所述待分发视频的视频相关信息；
根据所述视频相关信息，确定所述待分发视频的时效周期，所述时效周期用于表示所述待分发视频对应的有效时长；
根据所述待分发视频的首发时间和所述时效周期，确定所述待分发视频的时效结 束时间。
由于上述方法由装置执行，不依赖审核人员的时效性标注经验，因此，不受审核人员主观因素干扰，标注质量稳定，也无需对审核人员进行培训，节省了人力物力，同时也提升了视频时效的确定效率。
可选的，所述视频相关信息包括以下至少一种信息：标题、作者、语音、视频、字幕、人物和标题句法分析。
可选的，所述根据所述视频相关信息，确定所述待分发视频的时效周期，包括：
通过对所述标题进行文本分类确定第一周期；
通过对所述视频进行新闻识别确定第二周期；
通过对所述语音、所述字幕、所述人物以及所述标题句法分析中至少一项的关键词进行识别确定第三周期；
根据所述作者确定第四周期；
将所述第一周期、所述第二周期、所述第三周期和所述第四周期中时长最短的周期确定为所述待分发视频的时效周期。
由于视频相关信息中每个信息都可能对应一个有效时长，本申请提供了上述四种确定周期的方式，分别通过这四种方式确定出对应的周期后，参考四个周期最终确定待分发视频的时效周期，解决了用单一方式确定最终时效周期带来的准确度不高的问题。
可选的，所述通过对所述标题进行文本分类确定第一周期，包括：
通过快速文本分类模型对所述标题进行分类，得到所述标题对应的文本类别；
根据所述标题对应的文本类别和第一映射关系，确定所述第一周期，所述第一映射关系表征本文类别和周期的对应关系。
可选的，所述通过对所述视频进行新闻识别确定第二周期，包括：
通过神经网络模型对所述视频进行识别，得到所述视频对应的视频类型；
根据所述视频对应的视频类型和第二映射关系，确定所述第二周期，所述第二映射关系表征视频类型和周期的对应关系。
可选的，所述第二映射关系中的视频类型包括：新闻类；新闻类视频对应的周期最短。
可选的，所述通过对所述语音、所述字幕、所述人物以及所述标题句法分析中至少一项的关键词进行识别确定第三周期，包括：
通过语音识别技术将所述语音转化为第一文字信息；
通过图像文字识别技术将所述字幕转化为第二文字信息；
对所述人物、所述标题句法分析、所述第一文字信息以及所述第二文字信息中至少一项的关键词进行识别，得到至少一个关键词；
根据所述至少一个关键词和第三映射关系，确定所述至少一个关键词各自对应的周期；
将所述至少一个关键词各自对应的周期中时长最短的周期确定为所述第三周期。
可选的，所述根据所述作者确定第四周期，包括：
根据所述待分发视频的作者和第四映射关系，确定所述第四周期，所述第四映射关系表征作者和周期的对应关系。
根据第二方面，本申请提供一种视频分发时效确定装置，包括：
解析模块，用于对待分发视频进行解析，得到所述待分发视频的视频相关信息；
时效性策略模块，用于根据所述视频相关信息，确定所述待分发视频的时效周期，所述时效周期用于表示所述待分发视频对应的有效时长；
时效性决策模块，用于根据所述待分发视频的首发时间和所述时效周期，确定所述待分发视频的时效结束时间。
可选的，所述视频相关信息包括以下至少一种信息：标题、作者、语音、视频、字幕、人物和标题句法分析。
可选的，所述时效性策略模块包括：标题文本分类单元、新闻识别单元、关键词识别单元、作者识别单元和确定单元；
所述标题文本分类单元用于通过对所述标题进行文本分类确定第一周期；
所述新闻识别单元用于通过对所述视频进行新闻识别确定第二周期；
所述关键词识别单元用于通过对所述语音、所述字幕、所述人物以及所述标题句法分析中至少一项的关键词进行识别确定第三周期；
所述作者识别单元用于根据所述作者确定第四周期。
所述确定单元用于将所述第一周期、所述第二周期、所述第三周期和所述第四周期中时长最短的周期确定为所述待分发视频的时效周期。
可选的，所述标题文本分类单元具体用于：
通过快速文本分类模型对所述标题进行分类，得到所述标题对应的文本类别；
根据所述标题对应的文本类别和第一映射关系，确定所述第一周期，所述第一映射关系表征本文类别和周期的对应关系。
可选的，所述新闻识别单元具体用于：
通过神经网络模型对所述视频进行识别，得到所述视频对应的视频类型；
根据所述视频对应的视频类型和第二映射关系，确定所述第二周期，所述第二映射关系表征视频类型和周期的对应关系。
可选的，所述第二映射关系中的视频类型包括：新闻类；新闻类视频对应的周期最短。
可选的，所述关键词识别单元具体用于：
通过语音识别技术将所述语音转化为第一文字信息；
通过图像文字识别技术将所述字幕转化为第二文字信息；
对所述人物、所述标题句法分析、所述第一文字信息以及所述第二文字信息中至少一项的关键词进行识别，得到得到至少一个关键词；
根据所述至少一个关键词和第三映射关系，确定所述至少一个关键词各自对应的周期；
将所述至少一个关键词各自对应的周期中时长最短的周期确定为所述第三周期。
可选的，所述作者识别单元具体用于：
根据所述待分发视频的作者和第四映射关系，确定所述第四周期，所述第四映射关系表征作者和周期的对应关系。
根据第三方面，本申请提供一种电子设备，包括：
至少一个处理器；以及
与所述至少一个处理器通信连接的存储器；其中，
所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行上述视频分发时效确定方法。
根据第四方面，本申请提供一种存储有计算机指令的非瞬时计算机可读存储介质，所述计算机指令用于使所述计算机执行上述视频分发时效确定方法。
本申请提供的视频分发时效确定方法和装置。不依赖审核人员的时效性标注经验，因此，不受审核人员主观因素干扰，标注质量稳定，也无需对审核人员进行培训，节省了人力物力，同时也提升了视频时效的确定效率。
应当理解，本部分所描述的内容并非旨在标识本公开的实施例的关键或重要特征，也不用于限制本公开的范围。本公开的其它特征将通过以下的说明书而变得容易理解。
附图说明
附图用于更好地理解本方案，不构成对本申请的限定。其中：
图1为本申请提供的视频分发时效确定方法的实施例一的流程示意图；
图2为本申请提供的视频分发时效确定方法的实施例二的流程示意图；
图3为本申请提供的第一映射关系示意图；
图4为本申请提供的第二映射关系示意图；
图5为本申请提供的第三映射关系示意图；
图6为本申请提供的第四映射关系示意图；
图7为本申请提供的视频分发时效确定装置的实施例的结构示意图；
图8为本申请提供的电子设备的框图。
具体实施方式
以下结合附图对本申请的示范性实施例做出说明，其中包括本申请实施例的各种细节以助于理解，应当将它们认为仅仅是示范性的。因此，本领域普通技术人员应当认识到，可以对这里描述的实施例做出各种改变和修改，而不会背离本申请的范围和精神。同样，为了清楚和简明，以下的描述中省略了对公知功能和结构的描述。
信息流推荐系统推荐给用户的视频应当是不会让用户觉得有过期感的视频，为了达到这种推荐效果，现有技术中，采用人工方式针对每个视频进行时效性标注，推荐系统进一步根据人工标注的时效性完成视频的分发，比如：审核人员根据主观经验将某视频的时效性标注为强时效，推荐系统进而在该强时效对应的时长内分发该视频。
然而，采用人工标注的方式会存在如下问题：受主观因素干扰严重，不同的审核人员对同一视频的标注结果可能是不同的，标注质量不稳定；灵活性差，新视频资源的接入和标注标准的变化都需要重新培训审核人员，耗费大量人力物力；视频量级巨大，人工标注的方式效率很低。
考虑到现有技术存在的上述技术问题，本申请提供一种视频分发时效确定方法和装置，该时效确定装置可以集成在信息流推荐系统中，将待分发视频输入该装置后，该装置便可通过执行本申请提供的视频分发时效确定方法来确定待分发视频的时效，并将该时效传递给信息流推荐系统的用于分发视频的模块，由该模块完成视频的分发，整个过程不需要人工干预，可以解决现有技术存在的上述技术问题。
针对上面描述的时效确定装置执行本申请提供的时效确定方法来确定待分发视频的时效的过程，介绍如下，时效确定装置接收到待分发视频后，解析得到待分发视频的视频相关信息，比如：标题、作者、语音、视频和字幕等；由于这些视频相关信息中每个信息都可能对应一个周期，因此可以根据这些视频相关信息，确定多个周期， 然后取其中时长最短的周期作为待分发视频的时效周期，最后结合待分发视频的首发时间，确定出待分发视频的时效结束时间，并将该时效结束时间传递给分发视频的模块，以使该模块在该时效结束时间之前完成上述待分发视频的分发。和现有技术人工标注时效性相比，由于上述方法由装置执行，不依赖审核人员的时效性标注经验，因此，不受审核人员主观因素干扰，标注质量稳定，也无需对审核人员进行培训，节省了人力物力，同时也提升了视频时效的确定效率。
下面结合具体实施例对本申请提供的视频分发时效确定方法的实现方式进行详细说明。
实施例一
图1为本申请提供的视频分发时效确定方法的实施例一的流程示意图。本实施例提供的视频分发时效确定方法可由上文介绍的时效确定装置来执行。如图1所示，本实施例提供的视频分发时效确定方法，包括：
S101、对待分发视频进行解析，得到所述待分发视频的视频相关信息。
可选的，待分发视频的视频相关信息可包括以下至少一种信息：标题、作者、语音、视频、字幕、人物和标题句法分析。
一种可能的实现方式中，用户将待分发视频上传到时效确定装置后，该时效确定装置从待分发视频的文件名称解析得到待分发视频的标题和作者，从待分发视频中解析得到待分发视频的语音、字幕和视频，其中，解析得到的该视频可以为带语音和字幕的待分发视频本身，也可为不带语音和字幕的动画片段。在解析得到标题的基础上，采用自然语言处理(Natural Language Processing，简称NLP)对标题进行处理得到标题句法分析，在解析得到视频的基础上，采用神经网络对视频中的人物进行人脸识别得到人物。
可选的，标题、作者、人物和标题句法分析对应的文件格式为文本格式，比如：.txt、.doc.、.docx、.wps等。语音对应的文件格式为音频格式，比如：MP3、MIDI、WMA、VQF、AMR等。字幕对应的文件格式为图片格式，比如：bmp、jpg、png、tif、gif等。视频对应的文件格式为视频格式，比如：wmv、asf、asx等。
对视频进行解析得到上述各种信息的解析原理参见现有技术，本申请在此不再赘述。
S102、根据所述视频相关信息，确定所述待分发视频的时效周期，所述时效周期用于表示所述待分发视频对应的有效时长。
一种可实现的方式中，可根据S101解析得到的标题、作者、语音、视频、字幕、 人物和标题句法分析中每个信息确定出一个对应的周期，得到多个周期后，取其中时长最短的周期作为S102中待分发视频的时效周期。
具体的，可通过文本分类、新闻识别、关键词识别以及作者识别等方法确定上述每个信息对应的周期。
S103、根据所述待分发视频的首发时间和所述时效周期，确定所述待分发视频的时效结束时间。
需要说明的是：本步骤S103中的首发时间可以是在对待分发视频解析时得到，也可以是解析得到标题、作者、人物等信息后，从这些信息中提取搜索关键字，在搜索引擎中搜索该关键字查询得到。
下面举例说明：
假设待分发视频的首发时间为2019年10月1日，通过S102确定得到待分发视频的时效周期为2天，那么时效结束时间便是首发时间加2天，即2019年10月3日。
本实施例提供的视频分发时效确定方法，接收到用户上传的待分发视频后，对该待分发视频进行解析，得到待分发视频的视频相关信息，然后根据该视频相关信息，确定待分发视频的时效周期，最后根据待分发视频的首发时间和上述时效周期，确定待分发视频的时效结束时间，和现有技术人工标注时效性相比，由于上述方法由装置执行，不依赖审核人员的时效性标注经验，因此，不受审核人员主观因素干扰，标注质量稳定，也无需对审核人员进行培训，节省了人力物力，同时也提升了视频时效的确定效率。
实施例二
图2为本申请提供的视频分发时效确定方法的实施例二的流程示意图。如图2所示，本实施例提供的视频分发时效确定方法，包括：
S201、对待分发视频进行解析，得到所述待分发视频的视频相关信息。
上述视频相关信息包括以下至少一种信息：标题、作者、语音、视频、字幕、人物和标题句法分析。S201的实现方式可参见实施例一，本申请在此不再赘述。
下面通过步骤S202-S205介绍不同视频相关信息对应的周期的确定过程。
S202、通过对上述标题进行文本分类确定第一周期。
具体的，首先通过快速文本分类模型对所述标题进行分类，得到所述标题对应的文本类别。然后根据所述标题对应的文本类别和第一映射关系，确定所述第一周期，所述第一映射关系表征本文类别和周期的对应关系。
下面对本实现方式举例说明：
图3为第一映射关系示意图，参见图3所示，文本类别例如可以包括：情感、军事、政治、教育以及娱乐等。在S201解析得到待分发视频的标题后，将该标题输入快速文本分类模型，假设快速文本分类模型输出的文本类别为娱乐，图3所示第一映射关系中与娱乐匹配的为强时效，强时效的周期为3天，则可确定第一周期为3天。
S203、通过对上述视频进行新闻识别确定第二周期。
具体的，首先通过神经网络模型对所述视频进行识别，得到所述视频对应的视频类型。然后根据所述视频对应的视频类型和第二映射关系，确定所述第二周期，所述第二映射关系表征视频类型和周期的对应关系。
下面对本实现方式举例说明：
图4为第二映射关系示意图，参见图4所示，视频类型例如可以包括：综艺类、新闻类、天气预报类以及影视类等。在S201解析得到待分发视频的视频后，将该视频输入神经网络模型，假设神经网络模型输出的视频类型为新闻类，图4所示第二映射关系中与新闻类匹配的为超强时效，超强时效的周期为2天，则可确定第二周期为2天。
S204、通过对所述语音、所述字幕、所述人物以及所述标题句法分析中至少一项的关键词进行识别确定第三周期。
需要说明的是：本实现方式中，可通过对语音、字幕、人物以及标题句法分析中任意一项的关键词进行识别来确定第三周期，比如仅对语音的关键词进行识别来确定第三周期。也可通过对语音、字幕、人物以及标题句法分析中任意两项或者多项的关键词进行识别确定来第三周期，比如对语音、字幕、人物以及标题句法分析所有项的关键词进行识别来确定第三周期。由于不同的关键词对应了不同的周期，当识别到的关键词有多个时，会得到多个周期，可将多个周期中时长最短的作为第三周期。
具体的，首先通过语音识别技术将语音转化为第一文字信息，通过图像文字识别技术将字幕转化为第二文字信息，然后对人物、标题句法分析、第一文字信息以及第二文字信息中至少一项的关键词进行识别，从而得到至少一个关键词；然后根据识别到的至少一个关键词和第三映射关系，确定至少一个关键词各自对应的周期，将其中时长最短的周期确定为第三周期。
下面对本实现方式举例说明，
图5为第三映射关系示意图，参见图5所示，关键词例如可以包括：财经、汽车、美食、游戏、旅游、巴菲特、下跌等。在S201解析得到待分发视频的语音、字幕、人物和标题句法分析后，采用语音识别技术将语音转化为第一文字信息，采用图像文字 识别技术将字幕转化为第二文字信息，假设从第一文字信息识别到的关键词为财经，从第二文字信息识别到的关键词为汽车，从人物中识别到的关键词为巴菲特，从标题句法分析识别到的关键词为下跌。根据第三映射关系可以确定财经和下跌对应周期为2天，汽车和巴菲特对应的周期为3天。那么可以将时长最短的周期确定为第三周期，即2天。
需要说明的是：上例是以对语音、字幕、人物以及标题句法分析所有项进行关键词识别为例对第三周期的确定过程进行说明，如上文描述，还可以择其中一项或者多项来确定第三周期，本申请对此不再一一赘述。
S205、根据所述作者确定第四周期。
具体的，根据所述待分发视频的作者和第四映射关系，确定所述第四周期，所述第四映射关系表征作者和周期的对应关系。
下面对本实现方式举例说明：
图6为第一映射关系示意图，参见图6所示，作者例如可以包括：作者A、作者B、作者C和作者D，在S201解析得到待分发视频的作者后，先判断该作者是否在图6所述映射关系中，如果在，继续根据图6所示映射关系确定对应的周期，假设对待分发视频解析得到的作者为作者A，参见图6可确定作者A对应的周期为2天，那么第四周期则为2天。
需要说明的是：图3-图6中涉及的无时效指的是：待分发视频没有时效要求，在任意时长内分发均可。
S206、将所述第一周期、所述第二周期、所述第三周期和所述第四周期中时长最短的周期确定为所述待分发视频的时效周期。
下面举例说明：
参见上文举例，通过对标题进行文本分类确定的第一周期为3天；通过对视频进行新闻识别确定的第二周期为2天；通过对语音、字幕、人物以及标题句法分析中至少一项的关键词进行识别确定的第三周期为2天；根据作者确定的第四周期为2天。其中，时长最短的周期为2天，则可将待分发视频的时效周期确定为2天。
S207、根据所述待分发视频的首发时间和所述时效周期，确定所述待分发视频的时效结束时间。
S207的实现方式可参见上述实施例的S103，本申请在此不再赘述。
本实施例提供的视频分发时效确定方法，提供了根据不同视频相关信息确定对应周期的可实现方式，在确定了多个周期的基础上，参考该多个周期最终确定待分发视 频的时效周期，解决了用单一方式确定最终时周期带来的准确度不高的问题。
图7为本申请提供的视频分发时效确定装置700的实施例的结构示意图。如图7所示，本实施例提供的视频分发时效确定装置700，包括：
解析模块701，用于对待分发视频进行解析，得到所述待分发视频的视频相关信息；
时效性策略模块702，用于根据所述视频相关信息，确定所述待分发视频的时效周期，所述时效周期用于表示所述待分发视频对应的有效时长；
时效性决策模块703，用于根据所述待分发视频的首发时间和所述时效周期，确定所述待分发视频的时效结束时间。
可选的，所述视频相关信息包括以下至少一种信息：标题、作者、语音、视频、字幕、人物和标题句法分析。
可选的，所述时效性策略模块702包括：标题文本分类单元7021、新闻识别单元7022、关键词识别单元7023、作者识别单元7024和确定单元7025；
所述标题文本分类单元7021用于通过对所述标题进行文本分类确定第一周期；
所述新闻识别单元7022用于通过对所述视频进行新闻识别确定第二周期；
所述关键词识别单元7023用于通过对所述语音、所述字幕、所述人物以及所述标题句法分析中至少一项的关键词进行识别确定第三周期；
所述作者识别单元7024用于根据所述作者确定第四周期。
所述确定单元7025用于将所述第一周期、所述第二周期、所述第三周期和所述第四周期中时长最短的周期确定为所述待分发视频的时效周期。
可选的，所述标题文本分类单元具体7021用于：
通过快速文本分类模型对所述标题进行分类，得到所述标题对应的文本类别；
根据所述标题对应的文本类别和第一映射关系，确定所述第一周期，所述第一映射关系表征本文类别和周期的对应关系。
可选的，所述新闻识别单元7022具体用于：
通过神经网络模型对所述视频进行识别，得到所述视频对应的视频类型；
根据所述视频对应的视频类型和第二映射关系，确定所述第二周期，所述第二映射关系表征视频类型和周期的对应关系。
可选的，所述第二映射关系中的视频类型包括：新闻类；新闻类视频对应的周期最短。
可选的，所述关键词识别单元7023具体用于：
通过语音识别技术将所述语音转化为第一文字信息；
通过图像文字识别技术将所述字幕转化为第二文字信息；
对所述人物、所述标题句法分析、所述第一文字信息以及所述第二文字信息中至少一项的关键词进行识别，得到至少一个关键词；
根据所述至少一个关键词和第三映射关系，确定所述至少一个关键词各自对应的周期；
将所述至少一个关键词各自对应的周期中时长最短的周期确定为所述第三周期。
可选的，所述作者识别单元具体7024用于：
根据所述待分发视频的作者和第四映射关系，确定所述第四周期，所述第四映射关系表征作者和周期的对应关系。
本实施例提供的视频分发时效确定装置，可用于执行上述任一方法实施例中对应的步骤，其实现原理和技术效果类似，在此不再赘述。
根据本申请的实施例，本申请还提供了一种电子设备和一种可读存储介质。
如图8所示，是根据本申请实施例的视频分发时效确定方法的电子设备的框图。电子设备旨在表示各种形式的数字计算机，诸如，膝上型计算机、台式计算机、工作台、个人数字助理、服务器、刀片式服务器、大型计算机、和其它适合的计算机。电子设备还可以表示各种形式的移动装置，诸如，个人数字处理、蜂窝电话、智能电话、可穿戴设备和其它类似的计算装置。本文所示的部件、它们的连接和关系、以及它们的功能仅仅作为示例，并且不意在限制本文中描述的和/或者要求的本申请的实现。
如图8所示，该电子设备包括：一个或多个处理器801、存储器802，以及用于连接各部件的接口，包括高速接口和低速接口。各个部件利用不同的总线互相连接，并且可以被安装在公共主板上或者根据需要以其它方式安装。处理器可以对在电子设备内执行的指令进行处理，包括存储在存储器中或者存储器上以在外部输入/输出装置(诸如，耦合至接口的显示设备)上显示GUI的图形信息的指令。在其它实施方式中，若需要，可以将多个处理器和/或多条总线与多个存储器和多个存储器一起使用。同样，可以连接多个电子设备，各个设备提供部分必要的操作(例如，作为服务器阵列、一组刀片式服务器、或者多处理器系统)。图8中以一个处理器801为例。
存储器802即为本申请所提供的非瞬时计算机可读存储介质。其中，所述存储器存储有可由至少一个处理器执行的指令，以使所述至少一个处理器执行本申请所提供的视频分发时效确定方法。本申请的非瞬时计算机可读存储介质存储计算机指令，该计算机指令用于使计算机执行本申请所提供的视频分发时效确定方法。
存储器802作为一种非瞬时计算机可读存储介质，可用于存储非瞬时软件程序、非瞬时计算机可执行程序以及模块，如本申请实施例中的视频分发时效确定方法对应的程序指令/模块(例如，附图7所示的解析模块701、时效性策略模块702和时效性决策模块703)。处理器801通过运行存储在存储器802中的非瞬时软件程序、指令以及模块，从而执行服务器的各种功能应用以及数据处理，即实现上述方法实施例中的视频分发时效确定方法。
存储器802可以包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需要的应用程序；存储数据区可存储用于实现视频分发时效确定方法的电子设备的使用所创建的数据等。此外，存储器802可以包括高速随机存取存储器，还可以包括非瞬时存储器，例如至少一个磁盘存储器件、闪存器件、或其他非瞬时固态存储器件。在一些实施例中，存储器802可选包括相对于处理器801远程设置的存储器，这些远程存储器可以通过网络连接至用于实现视频分发时效确定方法的电子设备。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
视频分发时效确定方法的电子设备还可以包括：输入装置803和输出装置804。处理器801、存储器802、输入装置803和输出装置804可以通过总线或者其他方式连接，图8中以通过总线连接为例。
输入装置803可接收输入的数字或字符信息，以及产生与用于实现视频分发时效确定方法的电子设备的用户设置以及功能控制有关的键信号输入，例如触摸屏、小键盘、鼠标、轨迹板、触摸板、指示杆、一个或者多个鼠标按钮、轨迹球、操纵杆等输入装置。输出装置804可以包括显示设备、辅助照明装置(例如，LED)和触觉反馈装置(例如，振动电机)等。该显示设备可以包括但不限于，液晶显示器(LCD)、发光二极管(LED)显示器和等离子体显示器。在一些实施方式中，显示设备可以是触摸屏。
此处描述的系统和技术的各种实施方式可以在数字电子电路系统、集成电路系统、专用ASIC(专用集成电路)、计算机硬件、固件、软件、和/或它们的组合中实现。这些各种实施方式可以包括：实施在一个或者多个计算机程序中，该一个或者多个计算机程序可在包括至少一个可编程处理器的可编程系统上执行和/或解释，该可编程处理器可以是专用或者通用可编程处理器，可以从存储系统、至少一个输入装置、和至少一个输出装置接收数据和指令，并且将数据和指令传输至该存储系统、该至少一个输入装置、和该至少一个输出装置。
这些计算程序(也称作程序、软件、软件应用、或者代码)包括可编程处理器的机器指令，并且可以利用高级过程和/或面向对象的编程语言、和/或汇编/机器语言来实施这些计算程序。如本文使用的，术语“机器可读介质”和“计算机可读介质”指的是用于将机器指令和/或数据提供给可编程处理器的任何计算机程序产品、设备、和/或装置(例如，磁盘、光盘、存储器、可编程逻辑装置(PLD))，包括，接收作为机器可读信号的机器指令的机器可读介质。术语“机器可读信号”指的是用于将机器指令和/或数据提供给可编程处理器的任何信号。
为了提供与用户的交互，可以在计算机上实施此处描述的系统和技术，该计算机具有：用于向用户显示信息的显示装置(例如，CRT(阴极射线管)或者LCD(液晶显示器)监视器)；以及键盘和指向装置(例如，鼠标或者轨迹球)，用户可以通过该键盘和该指向装置来将输入提供给计算机。其它种类的装置还可以用于提供与用户的交互；例如，提供给用户的反馈可以是任何形式的传感反馈(例如，视觉反馈、听觉反馈、或者触觉反馈)；并且可以用任何形式(包括声输入、语音输入或者、触觉输入)来接收来自用户的输入。
可以将此处描述的系统和技术实施在包括后台部件的计算系统(例如，作为数据服务器)、或者包括中间件部件的计算系统(例如，应用服务器)、或者包括前端部件的计算系统(例如，具有图形用户界面或者网络浏览器的用户计算机，用户可以通过该图形用户界面或者该网络浏览器来与此处描述的系统和技术的实施方式交互)、或者包括这种后台部件、中间件部件、或者前端部件的任何组合的计算系统中。可以通过任何形式或者介质的数字数据通信(例如，通信网络)来将系统的部件相互连接。通信网络的示例包括：局域网(LAN)、广域网(WAN)和互联网。
计算机系统可以包括客户端和服务器。客户端和服务器一般远离彼此并且通常通过通信网络进行交互。通过在相应的计算机上运行并且彼此具有客户端-服务器关系的计算机程序来产生客户端和服务器的关系。
根据本申请实施例的技术方案，和现有技术人工标注时效性相比，由于本申请实施例的技术方案由装置执行，不依赖审核人员的时效性标注经验，因此，不受审核人员主观因素干扰，标注质量稳定，也无需对审核人员进行培训，节省了人力物力，同时也提升了视频时效的确定效率。
应该理解，可以使用上面所示的各种形式的流程，重新排序、增加或删除步骤。例如，本发申请中记载的各步骤可以并行地执行也可以顺序地执行也可以不同的次序执行，只要能够实现本申请公开的技术方案所期望的结果，本文在此不进行限制。
上述具体实施方式，并不构成对本申请保护范围的限制。本领域技术人员应该明白的是，根据设计要求和其他因素，可以进行各种修改、组合、子组合和替代。任何在本申请的精神和原则之内所作的修改、等同替换和改进等，均应包含在本申请保护范围之内。

Claims (11)

  1. 一种视频分发时效确定方法，应用于视频分发时效确定装置，其特征在于，包括：
    对待分发视频进行解析，得到所述待分发视频的视频相关信息；
    根据所述视频相关信息，确定所述待分发视频的时效周期，所述时效周期用于表示所述待分发视频对应的有效时长；
    根据所述待分发视频的首发时间和所述时效周期，确定所述待分发视频的时效结束时间。
  2. 根据权利要求1所述的方法，其特征在于，所述视频相关信息包括以下至少一种信息：标题、作者、语音、视频、字幕、人物和标题句法分析。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述视频相关信息，确定所述待分发视频的时效周期，包括：
    通过对所述标题进行文本分类确定第一周期；
    通过对所述视频进行新闻识别确定第二周期；
    通过对所述语音、所述字幕、所述人物以及所述标题句法分析中至少一项的关键词进行识别确定第三周期；
    根据所述作者确定第四周期；
    将所述第一周期、所述第二周期、所述第三周期和所述第四周期中时长最短的周期确定为所述待分发视频的时效周期。
  4. 根据权利要求3所述的方法，其特征在于，所述通过对所述标题进行文本分类确定第一周期，包括：
    通过快速文本分类模型对所述标题进行分类，得到所述标题对应的文本类别；
    根据所述标题对应的文本类别和第一映射关系，确定所述第一周期，所述第一映射关系表征本文类别和周期的对应关系。
  5. 根据权利要求3所述的方法，其特征在于，所述通过对所述视频进行新闻识别确定第二周期，包括：
    通过神经网络模型对所述视频进行识别，得到所述视频对应的视频类型；
    根据所述视频对应的视频类型和第二映射关系，确定所述第二周期，所述第二映射关系表征视频类型和周期的对应关系。
  6. 根据权利要求5所述的方法，其特征在于，所述第二映射关系中的视频类型包 括：新闻类；新闻类视频对应的周期最短。
  7. 根据权利要求3所述的方法，其特征在于，所述通过对所述语音、所述字幕、所述人物以及所述标题句法分析中至少一项的关键词进行识别确定第三周期，包括：
    通过语音识别技术将所述语音转化为第一文字信息；
    通过图像文字识别技术将所述字幕转化为第二文字信息；
    对所述人物、所述标题句法分析、所述第一文字信息以及所述第二文字信息中至少一项的关键词进行识别，得到至少一个关键词；
    根据所述至少一个关键词和第三映射关系，确定所述至少一个关键词各自对应的周期；
    将所述至少一个关键词各自对应的周期中时长最短的周期确定为所述第三周期。
  8. 根据权利要求3所述的方法，其特征在于，所述根据所述作者确定第四周期，包括：
    根据所述待分发视频的作者和第四映射关系，确定所述第四周期，所述第四映射关系表征作者和周期的对应关系。
  9. 一种视频分发时效确定装置，其特征在于，包括：
    解析模块，用于对待分发视频进行解析，得到所述待分发视频的视频相关信息；
    时效性策略模块，用于根据所述视频相关信息，确定所述待分发视频的时效周期，所述时效周期用于表示所述待分发视频对应的有效时长；
    时效性决策模块，用于根据所述待分发视频的首发时间和所述时效周期，确定所述待分发视频的时效结束时间。
  10. 一种电子设备，其特征在于，包括：
    至少一个处理器；以及
    与所述至少一个处理器通信连接的存储器；其中，
    所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行权利要求1-8中任一项所述的方法。
  11. 一种存储有计算机指令的非瞬时计算机可读存储介质，其特征在于，所述计算机指令用于使所述计算机执行权利要求1-8中任一项所述的方法。
PCT/CN2020/113970 2020-01-03 2020-09-08 视频分发时效的确定方法和装置 WO2021135358A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
KR1020217008845A KR102542232B1 (ko) 2020-01-03 2020-09-08 비디오 배포 시효 결정 방법 및 장치
EP20862000.5A EP3872651A4 (en) 2020-01-03 2020-09-08 METHOD AND DEVICE FOR DETERMINING TIMELINESS OF VIDEO DISTRIBUTION
JP2021517017A JP7140913B2 (ja) 2020-01-03 2020-09-08 映像配信時効の決定方法及び装置
US17/186,270 US20210192222A1 (en) 2020-01-03 2021-02-26 Video distribution timeliness determining method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010005825.8 2020-01-03
CN202010005825.8A CN111177462B (zh) 2020-01-03 2020-01-03 视频分发时效的确定方法和装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/186,270 Continuation US20210192222A1 (en) 2020-01-03 2021-02-26 Video distribution timeliness determining method and apparatus

Publications (1)

Publication Number Publication Date
WO2021135358A1 true WO2021135358A1 (zh) 2021-07-08

Family

ID=70646410

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/113970 WO2021135358A1 (zh) 2020-01-03 2020-09-08 视频分发时效的确定方法和装置

Country Status (4)

Country Link
EP (1) EP3872651A4 (zh)
JP (1) JP7140913B2 (zh)
CN (1) CN111177462B (zh)
WO (1) WO2021135358A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111177462B (zh) * 2020-01-03 2023-05-30 百度在线网络技术（北京）有限公司 视频分发时效的确定方法和装置
CN113407714B (zh) * 2020-11-04 2024-03-12 腾讯科技（深圳）有限公司 基于时效的数据处理方法、装置、电子设备及存储介质
CN114666655A (zh) * 2022-03-15 2022-06-24 北京达佳互联信息技术有限公司 视频处理方法、装置、电子设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103297327A (zh) * 2013-05-29 2013-09-11 深圳数字电视国家工程实验室股份有限公司 消息推送方法、装置及显示终端
CN109086345A (zh) * 2018-07-12 2018-12-25 北京奇艺世纪科技有限公司 一种内容识别方法、内容分发方法、装置及电子设备
CN109889864A (zh) * 2019-01-03 2019-06-14 百度在线网络技术（北京）有限公司 视频推送方法、装置、设备及存储介质
US20190220473A1 (en) * 2012-03-12 2019-07-18 Google Llc Dynamic display of content consumption by geographic location
CN110278447A (zh) * 2019-06-26 2019-09-24 北京字节跳动网络技术有限公司 基于连续特征的视频推送方法、装置及电子设备
CN111177462A (zh) * 2020-01-03 2020-05-19 百度在线网络技术（北京）有限公司 视频分发时效的确定方法和装置

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7336890B2 (en) * 2003-02-19 2008-02-26 Microsoft Corporation Automatic detection and segmentation of music videos in an audio/video stream
US20050193015A1 (en) * 2004-02-19 2005-09-01 Sandraic Logic, Llc A California Limited Liability Company Method and apparatus for organizing, sorting and navigating multimedia content
EP1842369B1 (en) * 2005-01-12 2020-04-08 Invidi Technologies Corporation Targeted impression model for broadcast network asset delivery
KR101227500B1 (ko) * 2006-05-17 2013-01-29 엘지전자 주식회사 디지털 방송 신호를 처리하는 장치 및 방법
US8494344B2 (en) * 2008-08-22 2013-07-23 Panasonic Corporation Recording and playback apparatus
US20150347415A1 (en) * 2014-05-30 2015-12-03 Apple Inc. Http live streaming dateranges
CN105022801B (zh) * 2015-06-30 2018-06-22 北京奇艺世纪科技有限公司 一种热门视频挖掘方法和装置
CN108829893B (zh) * 2018-06-29 2021-01-29 北京百度网讯科技有限公司 确定视频标签的方法、装置、存储介质和终端设备
CN109840291A (zh) * 2018-12-29 2019-06-04 网易传媒科技（北京）有限公司 视频数据处理方法和装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190220473A1 (en) * 2012-03-12 2019-07-18 Google Llc Dynamic display of content consumption by geographic location
CN103297327A (zh) * 2013-05-29 2013-09-11 深圳数字电视国家工程实验室股份有限公司 消息推送方法、装置及显示终端
CN109086345A (zh) * 2018-07-12 2018-12-25 北京奇艺世纪科技有限公司 一种内容识别方法、内容分发方法、装置及电子设备
CN109889864A (zh) * 2019-01-03 2019-06-14 百度在线网络技术（北京）有限公司 视频推送方法、装置、设备及存储介质
CN110278447A (zh) * 2019-06-26 2019-09-24 北京字节跳动网络技术有限公司 基于连续特征的视频推送方法、装置及电子设备
CN111177462A (zh) * 2020-01-03 2020-05-19 百度在线网络技术（北京）有限公司 视频分发时效的确定方法和装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3872651A4

Also Published As

Publication number Publication date
EP3872651A1 (en) 2021-09-01
JP7140913B2 (ja) 2022-09-21
CN111177462B (zh) 2023-05-30
EP3872651A4 (en) 2022-07-27
JP2022518645A (ja) 2022-03-16
CN111177462A (zh) 2020-05-19

Similar Documents

Publication Publication Date Title
WO2021135358A1 (zh) 视频分发时效的确定方法和装置
US9456229B2 (en) Parsing single source content for multi-channel publishing
US11508153B2 (en) Method for generating tag of video, electronic device, and storage medium
JP7334395B2 (ja) ビデオ分類方法、装置、機器、および記憶媒体
US20120158742A1 (en) Managing documents using weighted prevalence data for statements
JP7163440B2 (ja) テキストクエリ方法、装置、電子機器、記憶媒体及びコンピュータプログラム製品
US10425493B2 (en) Adjusting prominence of a participant profile in a social networking interface
AU2017216520A1 (en) Common data repository for improving transactional efficiencies of user interactions with a computing device
CN110990057B (zh) 小程序子链信息的提取方法、装置、设备及介质
CN107924398B (zh) 用于提供以评论为中心的新闻阅读器的系统和方法
US20200174996A1 (en) Visual data summaries with cognitive feedback
CN110532404B (zh) 一种源多媒体确定方法、装置、设备及存储介质
CN110245334B (zh) 用于输出信息的方法和装置
US10942954B2 (en) Dataset adaptation for high-performance in specific natural language processing tasks
CN112015866B (zh) 用于生成同义文本的方法、装置、电子设备及存储介质
CN112328749A (zh) 知识要素提取方法、装置、电子设备、介质和程序产品
US9893944B2 (en) Managing network bandwidth based on cognitive analysis of site content against organizational needs
CN111353070A (zh) 视频标题的处理方法、装置、电子设备及可读存储介质
US20210192222A1 (en) Video distribution timeliness determining method and apparatus
CN114218431A (zh) 视频搜索方法、装置、电子设备以及存储介质
CN108664535B (zh) 信息输出方法和装置
CN113742523B (zh) 文本核心实体的标注方法及装置
KR20080020011A (ko) 모바일 웹 콘텐츠 서비스 시스템 및 그 방법
CN113268987B (zh) 一种实体名称识别方法、装置、电子设备和存储介质
CN111291201B (zh) 一种多媒体内容分值处理方法、装置和电子设备

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2021517017

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2020862000

Country of ref document: EP

Effective date: 20210318

NENP Non-entry into the national phase

Ref country code: DE