WO2022247906A1 - 直播处理方法、直播平台、装置、系统、介质及设备 - Google Patents

直播处理方法、直播平台、装置、系统、介质及设备 Download PDF

Info

Publication number
WO2022247906A1
WO2022247906A1 PCT/CN2022/095233 CN2022095233W WO2022247906A1 WO 2022247906 A1 WO2022247906 A1 WO 2022247906A1 CN 2022095233 W CN2022095233 W CN 2022095233W WO 2022247906 A1 WO2022247906 A1 WO 2022247906A1
Authority
WO
WIPO (PCT)
Prior art keywords
live
live broadcast
marking
spectator
data
Prior art date
Application number
PCT/CN2022/095233
Other languages
English (en)
French (fr)
Inventor
刘伟科
陈丽蓉
张然
王永哲
韩卫召
沈俊杰
Original Assignee
北京沃东天骏信息技术有限公司
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 北京沃东天骏信息技术有限公司 filed Critical 北京沃东天骏信息技术有限公司
Publication of WO2022247906A1 publication Critical patent/WO2022247906A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/21Server components or server architectures
    • H04N21/218Source of audio or video content, e.g. local disk arrays
    • H04N21/2187Live feed
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4668Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/83Generation or processing of protective or descriptive data associated with content; Content structuring
    • H04N21/835Generation of protective data, e.g. certificates
    • H04N21/8352Generation of protective data, e.g. certificates involving content or source identification data, e.g. Unique Material Identifier [UMID]

Definitions

  • the embodiments of the present application relate to the field of computer technology, for example, to a live broadcast processing method, a live broadcast platform, a device, a system, a medium, and equipment.
  • the live broadcast platform There are generally three main participants on the live broadcast platform: the live broadcaster, the viewer, and the monetization channel.
  • the core link is that the live broadcaster gathers the viewers and then monetizes.
  • a common example is that various platforms provide live broadcast services, live broadcasters display products, and viewers watch the live broadcast, which improves the audience's awareness of the brand or facilitates the conversion of orders.
  • the live broadcast platform includes a large number of live broadcasts, and each live broadcast usually broadcasts dozens or even dozens of hundreds of products, and each product lasts for a few minutes. Time is very random. If users do not watch the whole process, they will easily miss the product display they want to see; if users watch the whole process, they will waste a lot of time on products that they do not pay attention to.
  • the embodiment of the present application provides a live broadcast processing method, live broadcast platform, device, system, medium and equipment, so as to realize accurate recommendation to the viewer end, improve the matching degree of the recommended marking video and the viewer end, and improve the accuracy of marking information.
  • the embodiment of the present application provides a live broadcast processing method, including:
  • the embodiment of the present application also provides a live broadcast processing method, including:
  • the embodiment of the present application also provides a live broadcast platform, including:
  • the recall request acquisition module is configured to obtain a recall request from the spectator end, wherein the spectator end is configured to mark the recommended live broadcast;
  • the current live field acquisition module is configured to obtain at least one current live field of live data
  • the recommended marking live broadcast determination module is configured to determine the recommended marking live broadcast of the spectator end based on the spectator field of the spectator end and the current live field of the at least one live data;
  • the marking information acquisition module is configured to send the live broadcast identification of the recommended marking live broadcast to the spectator end, and receive the marking information fed back by the spectator end, and store the marking information of the at least one live data, wherein , the marking information is recommended reference information corresponding to live data.
  • the embodiment of the present application also provides a live broadcast processing device, including:
  • the first data request module is configured to send a recall request to the live broadcast platform, and receive the live broadcast identification of the recommended marking live broadcast that matches the spectator field of the current device;
  • the second data request module is configured to generate a live data request based on the selected live broadcast identification of the recommended marking live broadcast, send the live data request to the live broadcast platform, and receive the live data fed back by the live broadcast platform and the live data to be marked information;
  • a marking module configured to display the live data and the information to be marked, and obtain the marking information corresponding to the information to be marked;
  • the marking information feedback module is configured to send the marking information and the live identification corresponding to the live data to the live broadcast platform, so that the live broadcast platform stores the marking information of the live data, wherein the The marking information is recommended reference information corresponding to the live data in the live broadcast platform.
  • the embodiment of the present application also provides a live broadcast system, including:
  • the live broadcast terminal is configured to obtain at least one live data, and send the at least one live data to the live platform;
  • Viewer set to send a recall request to the live broadcast platform
  • the live broadcast platform is configured to obtain the current live broadcast field of the at least one live data, determine the recommended marking live broadcast of the spectator end based on the spectator field and the current live broadcast field of the at least one live data, and mark the recommended live broadcast
  • the live broadcast logo of the marked live broadcast is sent to the viewer;
  • the spectator terminal broadcasts the recommended marking live broadcast, displays the recommended marking live broadcast, collects marking information of the recommended marking live broadcast, and sends the marking information to the live broadcast platform;
  • the live broadcast platform is further configured to receive the marking information fed back by the spectators, and store the marking information of the at least one live data, wherein the marking information is recommended reference information corresponding to the live data.
  • the embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor.
  • the processor executes the program, it implements the The live broadcast processing method provided by any embodiment.
  • the embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the live broadcast processing method provided in any embodiment of the present application is implemented.
  • FIG. 1 is a schematic flowchart of a live broadcast processing method provided in Embodiment 1 of the present application;
  • FIG. 2 is a schematic diagram of an application scenario provided by an embodiment of the present application.
  • FIG. 3 is a schematic flowchart of a live broadcast processing method provided in Embodiment 2 of the present application.
  • FIG. 4 is a schematic flowchart of a live broadcast processing method provided in Embodiment 3 of the present application.
  • FIG. 5 is a schematic structural diagram of a live broadcast platform provided in Embodiment 4 of the present application.
  • FIG. 6 is a schematic structural diagram of a live broadcast platform provided by an embodiment of the present application.
  • Fig. 7 is a schematic diagram of live room recommendation provided by the embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of a live broadcast processing device provided in Embodiment 5 of the present application.
  • FIG. 9 is a schematic structural diagram of a live broadcast system provided in Embodiment 6 of the present application.
  • FIG. 10 is a schematic structural diagram of an electronic device provided in Embodiment 7 of the present application.
  • FIG. 1 is a schematic flow diagram of a live broadcast processing method provided by Embodiment 1 of the present application. This embodiment can be applied to the situation where a large amount of live data is accurately recommended to the viewer who marks the live data. This method can be implemented by this application.
  • a live broadcast platform provided in the embodiment of the application is implemented, the live broadcast platform can be implemented by software and/or hardware, and the live broadcast platform can be configured on an electronic computing device, including the following steps:
  • S120 Acquire at least one current live broadcast field of live data, and determine the recommended marking live broadcast of the viewer terminal based on the viewer field and the current live broadcast field of the at least one live data.
  • FIG. 2 is a schematic diagram of an application scenario provided by an embodiment of the present application.
  • the live broadcast terminal, the customer terminal and the client end are respectively connected to the live broadcast platform.
  • the live broadcast terminal, the customer terminal and the user terminal can be three terminals respectively, that is, three different electronic devices; in other embodiments , it can also be that any two or three of the live broadcast terminal, spectator terminal and user terminal are the same electronic device, and different user permissions can be distinguished through different display interfaces, or different application programs (Application, APP) can be logged in. Distinguish, or perform regional verification by verifying different role identities.
  • Application Application
  • the live broadcast terminal, the viewer terminal and the client terminal can be terminal devices such as mobile phones and tablet computers, which have different permissions.
  • the live broadcast terminal is set to provide live broadcast data, and has the live background management authority, live broadcast object setting rights and live broadcast effect report query authority.
  • the viewer can watch the live broadcast and comment on the live broadcast, and has the right to evaluate the live broadcast and the platform.
  • the client can watch the live broadcast, and only has the right to watch the live broadcast. Among them, both the spectator and the client have the transaction authority to the live broadcast objects in the live broadcast.
  • the live broadcast terminal and the viewer terminal need to register respectively.
  • the live broadcast platform is equipped with a registration module, which receives the user's registration request and user information, registers the user's live broadcast identity or audience identity, and has corresponding permissions after successful registration.
  • a registration module which receives the user's registration request and user information, registers the user's live broadcast identity or audience identity, and has corresponding permissions after successful registration.
  • the live broadcast end you can set up a live broadcast, set the live broadcast object (such as the item or game to be introduced), generate live data, and transmit the generated live data to the live broadcast platform in real time, so that the viewer and the client can obtain the live data from the live broadcast platform.
  • the proportion of real traffic is increased, the proportion of virtual traffic is correspondingly reduced, and the situation of malicious swiping of traffic is reduced.
  • the spectator field of the spectator terminal is determined, and the spectator field of the spectator terminal is the specialty field of the user of the spectator terminal.
  • the recall request of the spectator terminal can be triggered when the spectator terminal successfully logs in, or is triggered when the spectator terminal detects a refresh operation.
  • the recall request of the spectator terminal includes the spectator identifier of the spectator terminal, for example, the ID of the spectator terminal, or the identifier of the user operating the spectator terminal.
  • determining the viewer based on the viewer ID in the recall request end of the field of spectators includes: matching the audience identifier of the audience terminal in a preset audience domain library, and determining at least one audience domain corresponding to the audience identifier.
  • the field classification logo is preset in the spectator field library, that is, different live broadcast fields correspond to different field classification logos, and the field classification logo can be updated according to changes in the live broadcast field, for example, adding a field classification logo or modifying the field classification logo Wait.
  • the spectator domain library stores the spectator ID and the domain classification ID corresponding to the spectator ID.
  • the spectator domain storehouse preferably stores the storage data structure of spectator identification-domain identification, and the storage data structure can be a redis-set structure, key is the field classification identification id, and value is the spectator identification (such as spectator id ) collection, the storage data structure can also be a redis-zset structure, key is the spectator identification (such as spectator id), value is the field classification identification id set, and score is the authoritative score value of the spectator in this category.
  • Data is stored through different structures, and fast queries can be performed based on different index information.
  • the field classification identifier id can be used to query the set of spectator IDs under the field, and in the stored data of the redis-zset structure, the field that the corresponding spectator is good at can be queried through the spectator ID A collection of taxonomies.
  • the field classification identifier corresponding to each spectator terminal in the spectator field database can be determined in two ways. One is that when the spectator terminal logs in to the live broadcast platform for the first time, the live broadcast platform sends the field classification information to the spectator terminal, and the spectator terminal collects the preset information selected by the user. Set the number (for example, two) of domain classification information, and send the domain classification identification of the selected domain classification information to the live broadcast platform, and the live broadcast platform updates the audience domain library based on the received audience identification and domain classification identification, and the audience terminal The authority score value of domain classification identification is recorded as zero.
  • the number of domain classification identifiers corresponding to different audience identifiers in the audience domain library may be different, for example, two or three, etc., which is not limited. In some optional embodiments, it may be based on the determination method of the domain classification identification corresponding to the customer terminal, and set the identification mark for the customer identification.
  • the determination method of the domain classification identification corresponding to the customer terminal includes the input of the customer terminal and marking evaluation , the identification identifier input by the spectator terminal can be 1, and the identification identifier identified by the marking evaluation can be 2.
  • the recall request from the spectator terminal when the recall request from the spectator terminal is received, match the spectator ID in the storage data of the redis-set structure based on the spectator ID in the recall request to obtain the field classification ID set corresponding to the spectator ID, based on The field classification identifiers in this set determine the audience field, and correspondingly, the query result in the audience field database can be 1 or 2.
  • the spectator ID fails to match in the spectator domain database, that is, the query result in the spectator domain database can be 0, indicating that the domain classification identifier corresponding to the spectator ID is not stored in the spectator domain database, and the field classification information is sent to the spectator terminal, and the spectator is received The spectator domain of terminal feedback, and update the spectator domain library.
  • the live broadcast data matching the field of viewers can be used as the recommended marking live broadcast for the corresponding audience.
  • the live broadcast field of the live broadcast data may be unique, and may change according to the change of the live broadcast object.
  • the current live field of the live data is determined based on the core live objects in the live data, wherein the core live objects can be set by the live end, for example, it can be a live broadcast whose duration is longer than half of the total duration of the live broadcast object.
  • the core live objects can be set by the live end, for example, it can be a live broadcast whose duration is longer than half of the total duration of the live broadcast object.
  • Any live object (such as commodities, goods, or online games, etc.) is configured with at least one level of category information, and the field classification identifier of the live object can be determined based on the minimum level of category information.
  • the category information of facial tissue includes daily necessities/paper products/facial tissue.
  • the levels of the above category information are descended in turn, and the domain classification identifier is determined according to the minimum level of category information, namely facial tissue, for example, the domain classification corresponding to facial tissue
  • the logo can be Z-01, and there is no limitation on the form of field classification logo.
  • obtaining the current live broadcast area of each live data includes: determining the current live broadcast area of the live data based on the current live object in the live data.
  • the same live broadcast can include multiple live broadcast objects, and the type of live broadcast objects is not limited. Therefore, in the same live broadcast data, different time periods can correspond to different live broadcast objects.
  • the live broadcast field corresponding to the live broadcast data also changes.
  • the previous live broadcast object is a mobile phone
  • the next live broadcast object is quick-frozen dumplings.
  • the live broadcast field changes with the live broadcast object. Detect the switch of the live broadcast object in the live broadcast data.
  • the current live broadcast field of the live broadcast data is updated based on the switched live broadcast object, for example, the minimum level category information of the live broadcast object after the switch is used to determine the live data.
  • the current live field may be to monitor the switching of the live object by identifying voice keywords in the live data, or to identify the currently displayed live object by performing image recognition on the image in the live data, or to monitor the switching of the live object.
  • the switching monitoring of the live broadcast object is realized through the monitoring of the object interface in the live broadcast data.
  • the object interface is the interface for accessing the live broadcast object, such as the details page access interface, the purchase interface, the current live product interface of the live broadcast page, etc.
  • the field database of live data can be changed in real time. When new live data is detected, live broadcast ends, or the current live field of live data changes, the live broadcast is updated. Domain database of data. Match the field of viewers in the field database of the live data to determine the recommended live broadcast of each spectator. Exemplarily, match at least one field of the spectator in the field database of the live data to determine at least one spectator For the live data corresponding to the field, the recommended marking live broadcast is determined in the live data corresponding to at least one spectator field.
  • the recommended marking live broadcast matches the fields that the viewer end users are good at, which is conducive to the fast and accurate marking of the recommended marking live broadcast by the viewer end, and improves the rate of recommending live broadcast data to the viewer end. Pertinence and accuracy improve the efficiency and accuracy of the marking information obtained from live data.
  • the live data can be improved. Recommendation specificity and precision.
  • the recommended marking live broadcast on the spectator side includes: the first live broadcast corresponding to the spectator field and the second live broadcast classified in other fields other than the spectator field, the first live broadcast and the second live broadcast The number of live broadcasts satisfies the preset ratio.
  • the first live broadcast can be determined based on the match between the viewer field of the spectator terminal and the current live broadcast field of the live broadcast data; Randomly determined from the live broadcast data corresponding to the field.
  • the quantity of the first live broadcast and the second live broadcast satisfies a preset ratio
  • the preset ratio may be between 1:1-5:1, and the preset ratio may be adjusted according to user requirements.
  • the preset ratio satisfied by the first live broadcast and the second live broadcast is determined based on the spectator field identification method of the spectator terminal, wherein the spectator field identification method includes spectator terminal input and marking evaluation identification .
  • the preset ratio between the first live broadcast and the second live broadcast may be 1:1; if the spectator field identification method is marking evaluation identification, then the first live broadcast and the second live broadcast
  • the preset ratio of the second live broadcast may be 3:1.
  • the live broadcast platform receives the access request sent by the viewer based on the recommended marking live broadcast, and feeds back the live data and corresponding marking information to the viewer.
  • sending the live data corresponding to the access request and the information to be marked corresponding to the live data to the spectator end includes: sending the live data corresponding to the access request to the watch the customer terminal, and perform live display timing; if the live display timing meets the preset display time, then send the information to be marked corresponding to the live data to the spectator terminal.
  • sending the live data corresponding to the access request and the information to be marked corresponding to the live data to the spectator end includes: synchronously sending the live data and the corresponding information to be marked to Viewer terminal to display the live broadcast data to the viewer terminal, and display the information to be marked when the live broadcast display time meets the preset display time.
  • the spectator terminal can display and mark the next live broadcast data after displaying the live data for a preset duration and marking, so that the traffic of the same spectator terminal can serve multiple live broadcast rooms , providing a new live viewing mode, which is conducive to breaking the traffic bottleneck caused by the fixed total number of Internet groups.
  • the live broadcast terminal is also set to set part of the message to be marked, and send this part of the message to be marked to the live broadcast platform, and the live broadcast platform will receive the part of the message to be marked sent by the live broadcast terminal, and another part of the pre-set
  • the message to be marked forms the message to be marked corresponding to the live data of the live broadcast terminal.
  • the message to be marked may be associated with a live broadcast identifier (such as a live broadcast ID (ie, a live broadcast room ID) or a live broadcast terminal ID).
  • the message to be marked can include but not limited to whether the food is attractive, whether the anchor tastes the effect satisfactorily, whether the anchor's make-up is obvious after eating, etc.
  • Another part of the message to be marked can be whether the current live broadcast is suspected of violating regulations, Whether the current live broadcast room is on-hook (the anchor is not in front of the camera), what is the current live broadcast sales category (clothes/food).
  • the message to be marked can be set according to the type of live broadcast and the live broadcast object, which is not limited.
  • the message to be marked corresponding to the above-mentioned live identification can be stored in a marking library, so that the corresponding message to be marked can be called according to the live identification.
  • the information to be marked sent to the spectator end includes information to be marked with known marks and information to be marked with unknown marks; wherein, the information to be marked with known marks can be Marked information to be marked, the known mark of the information to be marked can be obtained by identifying the live screen, exemplary, the known marked information to be marked can be but not limited to the number of live broadcasters in the screen, the anchor Whether it is in the picture or not.
  • the successfully verified marking information count the marking information fed back by at least one spectator terminal, and determine the target marking information in the marking information, and verify the marking fed back by the at least one spectator terminal based on the target marking information correctness of information.
  • the preset ratio can be 70%, the preset number can be 500 or 1000, etc., which can be adjusted according to requirements.
  • early warning prompt information is generated, and the early warning prompt information is sent to the live broadcast terminal, or the corresponding live broadcast data is blocked.
  • the correctness of the marking information fed back by the at least one spectator terminal is verified based on the target marking information, that is, for the same information to be marked, marking information different from the target marking information is determined to be incorrect.
  • a marking log is formed for the correct marking information, which is used to record the client identification, live broadcast identification, marking time stamp, information to be marked and marking information.
  • the marking message of at least one live data is stored, and the marking message is used as recommendation reference information corresponding to the video data, so that the pertinence of recommendation for at least one live data is improved.
  • the method further includes: evaluating the marking information, and determining the evaluation result of the marking information; based on at least one field
  • the evaluation result of the marking information in the classification updates the spectator field of the spectator terminal in the spectator field database.
  • evaluating the marking information may be to determine the correct rate and marking efficiency of the marking information, that is, the evaluation results include but not limited to the correct rate of marking and the marking efficiency. For any field classification, based on the evaluation results of the historical marking information corresponding to the field classification and the number of markings to determine the degree of proficiency in the field classification, based on the ranking of the proficiency in different field classifications, determine the current spectator field of the spectator terminal, and Update the spectator field library.
  • the spectator domain database may be updated based on a preset time interval, for example, the preset time interval may be one day, two days, or one week.
  • the live broadcast of the recommended marking live by determining the recommended marking live broadcast of the spectator end based on the field of viewers and the current live field of the at least one live data in the at least one current live data, the live broadcast of the recommended marking live
  • the identification is sent to the spectator end, so that the spectator end marks the recommended marking live broadcast. Since the recommended marking live broadcast is determined according to the field of viewers, within the field of expertise of the audience, it is beneficial for the audience to mark the live data in the field they are good at, and improve the accuracy and efficiency of marking information.
  • the marking information of at least one live data is stored, so that when the client requests live data, accurate recommendations can be made to the client based on the marking information, and the accuracy of live data recommendation can be improved.
  • the method further includes: setting a mark on the at least one live data based on the marking information of at least one spectator terminal, and updating the mark corresponding to the at least one live data to the marking library; receiving The recall request sent by the client, based on the mark and/or user portrait in the recall request, and the mark of at least one live data in the marking library, determine the recommended live data, and send the recommended live data to the client.
  • At least one target marking information will be determined to form a mark corresponding to the live broadcast identification. For example, if the information to be marked is whether the host is speaking too fast, if the target marking information is yes, then it is determined that the marking is too fast. At least one mark corresponding to each live broadcast identifier is updated and stored in the marking library, and the marking library saves the mark of each live broadcast through redis.
  • the recall request may be triggered by starting the platform application, or by refreshing the platform application, or by searching the platform application.
  • the recall request may include at least one item of a search mark and a user portrait.
  • the live broadcast platform may preset a corresponding relationship between user portraits and tags, and the corresponding tags may be determined based on the user portraits.
  • Match the mark in the recall request or the mark determined by the user portrait in the marking library determine at least one live identification corresponding to the mark, determine the live data corresponding to the live identification as recommended live data, and recommend it to the client.
  • a watcher may act as a client.
  • the live data is actually marked, and the corresponding live data is matched for the client through the mark, which improves the accuracy of matching and the recommendation accuracy of live broadcast, which is conducive to improving the conversion rate of live broadcast.
  • the technical solution of this embodiment avoids the problem of fragmentation of the live broadcast experience by adding a spectator terminal to closely integrate the live broadcast data with the viewer.
  • Real-time online marking of live broadcast data to realize real-time perception of live broadcast content which is convenient for the live broadcast platform to obtain the real status of at least one live broadcast in real time and accurately locate at least one live broadcast passing mark.
  • the tag matches the recommended live data with high accuracy, avoiding the situation of displaying a large amount of live data that users do not care about.
  • the method further includes: generating a mark based on the at least one live data, the mark of at least one live data, and the spectator terminal corresponding to the mark Reporting, sending the marking report to the corresponding live broadcast terminal, wherein the marking report includes at least one item of live marking results, audience marking results, live broadcast-tagging relationship, and marking audience contributions.
  • the live marking result is used to query the diagnosis result of the current live broadcast through the live broadcast id, which is the data cache of the redis-zset structure
  • the key is the live broadcast id
  • the value is the marking set
  • the score is the marking time.
  • live broadcast id-123456 marking collection: ⁇ value: "Speaking too fast”, socre: 1650151159, value: "Damning lights”, score: 1590123123 ⁇ .
  • the spectator marking result is used to query the spectator's marking result through the spectator id, which is a data cache of the redis-zset structure.
  • the key is the viewer ID
  • the hash-key is the live broadcast ID
  • the hash-value is the marking result.
  • watch client a mark the live broadcast id-123456, and mark the set: ⁇ "Speech too fast", "Lighting is normal" ⁇ .
  • the live broadcast-tag relationship is used to query the live broadcast list through tags, and it is a data cache of the redis-zset structure.
  • the key is the tag, the value is the live id, and the score is the timestamp. Such as tag-clothes, value- ⁇ id:123, score:1505454122 ⁇ .
  • Mark viewer contribution is used to count the contribution of viewers after the live broadcast ends.
  • key is a mark
  • set-key is a live broadcast id
  • list is a list of spectator ids.
  • mark-wool coat set-key- ⁇ id: 1, list: [viewer id: 1, viewer id: 2] ⁇ .
  • the live broadcast user can query the mark and improve it through marking, which is conducive to improving the quality of the live broadcast.
  • a hot tag-live cache data is maintained through the real-time updated marking library, which is used to store the corresponding relationship between at least one tag and at least one corresponding live broadcast in the current preset time interval, wherein a tag can be Corresponding to at least one live broadcast identifier, any live broadcast identifier may be associated with at least one mark.
  • the current preset time interval may be a preset time interval before the current moment, for example, 5 minutes. Markers in marker-live cache data can be sorted based on updated timestamps. It should be noted that the marker-live cache data may be updated in real time, or based on a fixed time interval.
  • setting marker-live cache data within a preset time interval includes: based on the At least one mark in the marking library, the time stamp of at least one mark and the live data corresponding to at least one mark, obtain the mark-live cache data in the current preset time interval; create a data update timed task, wherein the data update The timing task is used to add the corresponding relationship between the tag and the live data in the newly added duration in the tag-live cache data, and remove the corresponding relationship between the tag and the live data in the expired duration in the tag-live cache data; mark, execute the data update scheduled task to update the mark-live cache data.
  • the newly added duration can be the time interval of the scheduled task, such as 10s
  • the tag in the newly added duration can be an incremental tag between adjacent scheduled tasks, that is, a tag received within 10s is added, and the tag is set with the
  • the marked live data establishes a corresponding relationship, that is, the corresponding relationship is set between the marked and the live identification of the live data for which the marked is set.
  • the expiration time is the fixed time that is farthest from the preset time interval from the current time, for example, 10s farthest from the current time, and the added time is the same as the expired time.
  • the tag for query is determined, the above tag is matched in the tag-live cache data, and the live data that matches successfully is determined as the recommended live data.
  • the latest tag is stored in the tag-live cache data, it reflects the latest experience of the viewing user.
  • the recommended live data is determined based on the latest tag matching, which is in line with the recommendations of a large number of viewing users and improves the recommendation of recommended live data. precision.
  • FIG. 3 is a schematic flow chart of a live broadcast processing method provided in Embodiment 2 of the present application.
  • the embodiment of the present application can be combined with various alternative solutions in the foregoing embodiments.
  • determining the recommended marking live broadcast of the spectator terminal based on the spectator domain and the current live broadcast domain of the at least one live data includes: according to the live broadcast status of the at least one live data, Determining at least one recommendable live broadcast data; determining the recommended marking live broadcast of the viewer terminal based on the viewer field and the at least one recommendable live broadcast data.
  • the method of the embodiment of the present application includes the following steps:
  • S220 Acquire a current live field of at least one live data, and determine at least one recommendable live data according to the live state of the at least one live data.
  • S240 Send the live broadcast identification of the recommended marking live broadcast to the spectator end, receive the marking information fed back by the spectator end, and store at least one live data marking information, wherein the marking information corresponds to Recommended reference information for live data.
  • the live broadcast status of the live data may be determined according to the duration of the live broadcast or the amount of marking information collected.
  • the live broadcast status may include marking in progress and marking completed.
  • the live broadcast data whose live broadcast status is marking can be determined as recommendable live broadcast data.
  • the purpose of determining the recommended marking live broadcast is to quickly determine the live broadcast content of the live broadcast data through the marking of the live broadcast data by the viewer end, so as to accurately recommend to the client end.
  • the live broadcast status of the live broadcast data is marking; if the live broadcast duration of the live broadcast object exceeds the preset duration or the number of marking information received for the live broadcast object exceeds the preset number , switch the live broadcast status of the live broadcast data to complete marking, wherein the preset duration can be a fixed duration, such as 5 minutes, or it can also be determined according to the total duration of the live broadcast object, and the preset duration can be the duration of the live broadcast object Half of the total live time.
  • the live broadcast status for any live broadcast object is at the end of the live broadcast or has received more than a specified number of marking information, it indicates that the marking for the live broadcast object has been completed, and there is a waste of marking and recommending the live broadcast data
  • the live data can be determined as non-recommended live data, and when the live data switches the live broadcast object, the live data can be recommended as live data.
  • the recommended live broadcast of marking is determined in the recommendable live data, and on the basis of ensuring the field pertinence of the recommended live broadcast of marking, it avoids invalid marking recommendation and the loss of viewer-side resources. waste, improving the effectiveness and accuracy of the recommended marking live broadcast.
  • the live broadcast status includes the live broadcast duration of the current live broadcast object, and the number of the recommended marking live broadcast is at least one.
  • the live broadcast status determines at least one recommendation priority of the recommended marking live broadcast, and sends the live broadcast identification of the recommended marking live broadcast to the viewer terminal based on the recommendation priority.
  • Fig. 4 is a schematic flow chart of a live broadcast processing method provided in Embodiment 3 of the present application. This embodiment can be applied to the situation where spectators mark live broadcasts recommended by the live broadcast platform.
  • the live processing device can be implemented by a live processing device, which can be implemented by software and/or hardware.
  • the live processing device can be configured on the spectator end, and the spectator end can be such as a mobile phone, a tablet computer or a personal computer (Personal Computer, PC) etc. electronic equipment.
  • the method includes:
  • S310 Send a recall request to the live broadcast platform, and receive a live broadcast identification of a recommended marking live broadcast that matches the spectator field of the current device.
  • S320 Generate a live data request based on the selected live broadcast identification recommended for marking live broadcast, send the live data request to the live broadcast platform, and receive live data and marking information fed back by the live broadcast platform.
  • the spectator side After the spectator role is successfully logged in, the spectator side either detects the refresh operation input by the user, or generates a recall request according to the preset refresh time interval, and sends the recall request to the live broadcast platform.
  • the recall request includes the viewer ID of the spectator terminal, and the spectator ID may be the unique identifier of the spectator terminal or the unique identifier of the user operating the spectator terminal.
  • the live broadcast platform can use the viewer ID to match in the audience field library to match the audience field corresponding to the audience logo, and determine the matching audience field according to the audience field to recommend marking live broadcast, and feed back the live broadcast logo of the recommended marking live broadcast to Look at the client.
  • the method further includes: receiving field classification information fed back by the live broadcast platform, and displaying the field classification information on a display interface; detecting user input The selection information of the field classification information, and the selected field classification information is sent to the live broadcast platform, wherein the selected field classification information is used to determine the audience field.
  • the live broadcast platform does not match the spectator field based on the spectator ID, it indicates that the spectator field library does not include the spectator ID and the spectator field corresponding to the spectator ID.
  • the spectator terminal receives and displays domain classification information, and the displayed domain classification information can be displayed differently through user input selection operations.
  • the selection marks are set before and after for differential display, wherein the selection marks can be dots or asterisks, etc., which are not limited.
  • the user can select a preset number of domain categories, for example, two.
  • the display interface of the field classification information may include an "OK" control, and after detecting that the "OK" control is triggered, feedback information is generated based on the selected field classification, and the feedback information includes the selected field classification ID and viewer ID, send the feedback information to the live broadcast platform, so that the live broadcast platform will update the audience ID and domain classification ID in the feedback information to the audience domain database, and determine the audience domain based on the domain classification ID.
  • the recommended marking live broadcast matches the fields that the viewer-side users are good at, which is conducive to the fast and accurate marking of the recommended marking live broadcast by the viewer-end, improving the pertinence and accuracy of recommending live data to the viewer-end, and improving the acquisition of live data The efficiency and accuracy of marking information.
  • the viewer terminal receives the live broadcast identification of the recommended marking live broadcast, wherein the recommended marking live broadcast includes: the first live broadcast corresponding to the audience domain and the second live broadcast classified in other fields other than the audience domain, and the number of the first live broadcast and the second live broadcast satisfies Default scale.
  • the viewer displays the recommended live broadcast of marking based on the live display, wherein the number of the recommended live broadcast of marking is at least one, and the display order of at least one recommended live broadcast of marking can be based on the live broadcast platform's recommendation of at least one recommended live broadcast of marking The order is determined.
  • the recommended marking live broadcast displayed on the spectator terminal may display the cover image and description information of the recommended marking live broadcast. cover.
  • the description information may include the description information of the host and/or the live broadcast object, so that the user can learn the relevant information of the recommended marking live broadcast through the displayed content.
  • the viewer detects the user's selection operation on at least one of the displayed recommended marking live broadcasts. Exemplarily, it may detect a click operation on at least one recommended marking live broadcast, such as a touch click operation or a click operation input through a mouse or keyboard. .
  • a live broadcast data request is generated based on the live broadcast identifier of the recommended marking live broadcast corresponding to the click operation, and sent to the live broadcast platform.
  • the live broadcast data request includes the live broadcast identifier and the viewer identifier.
  • the live broadcast platform receives the live data request, matches the corresponding live data and the corresponding information to be marked based on the live broadcast identifier, and feeds back the matched live data and the information to be marked to the viewer.
  • the spectator terminal detects the selection operation of any of the displayed recommended marking live broadcasts, it enters the live broadcast display interface, receives the live broadcast data fed back by the live broadcast platform, and displays the live broadcast data on the live broadcast display interface.
  • the live data is displayed, and live display timing is performed, and if the live display timing meets a preset display duration, the information to be marked is displayed.
  • the viewer can receive the live data and the corresponding information to be marked synchronously. It can also receive the live data first, display the live data, and perform live display timing. If the live display timing If the preset display time is satisfied, a trigger signal is sent to the live broadcast platform, the information to be marked is received from the live broadcast platform, and the information to be marked is displayed.
  • the information to be marked may include a plurality of questions, and each question may correspond to at least two options for the user to choose, or correspond to an information input position, which is convenient for the user to input the marking information.
  • the spectator detection is used to input the information to be marked on the display interface, such as the selection operation of any option of the information to be marked, the selected option as the marking information, or the input operation at the information input position, the input
  • the content is used as marking information, etc.
  • the spectator terminal when the spectator terminal sends a recall request to the live broadcast platform, it receives the live broadcast platform’s feedback of the recommended marking live broadcast that matches the spectator field of the spectator terminal, and can request the selected live broadcast data from the live broadcast platform according to the user’s choice. , and display the live data and mark the live data based on the information to be marked. Since the recommended live broadcast of marking matches the fields that the spectator users are good at, it is beneficial for the spectator to quickly and accurately mark the recommended live broadcast Carry out marking, improve the pertinence and accuracy of recommending live data to the audience, and improve the efficiency and accuracy of marking information obtained from live data.
  • FIG. 5 is a schematic structural diagram of a live broadcast platform provided in Embodiment 4 of the present application.
  • the live broadcast platform includes:
  • the recall request acquisition module 410 is configured to obtain a recall request from the spectator end, wherein the spectator end is configured to mark the recommended live broadcast;
  • the current live field acquisition module 420 is configured to obtain at least one current live field of live data
  • Recommend marking live broadcast determination module 430 set to determine the recommended marking live broadcast of the spectator end based on the current live broadcast field of the spectator field and the at least one live data;
  • the marking information acquisition module 440 is configured to send the live broadcast identification of the recommended marking live broadcast to the spectator terminal, and receive the marking information fed back by the spectator terminal, and store the marking information of the at least one live data, Wherein, the marking information is recommended reference information corresponding to live data.
  • the recall request acquisition module 410 is set to:
  • the spectator field database does not include the spectator identifier, send field classification information to the spectator terminal, receive the spectator field fed back by the spectator terminal, and update the spectator field database.
  • the current live field acquisition module 420 is set to:
  • any live data determine the current live domain of any live data based on the core live objects in any live data, or determine any live data based on the current live objects in any live data The current live domain of the data.
  • the live broadcast recommended by the spectator end includes: the first live broadcast corresponding to the spectator field and the second live broadcast classified in other fields other than the spectator field, the first live broadcast and the The number of the second live broadcast satisfies a preset ratio.
  • the preset ratios satisfied by the first live broadcast and the second live broadcast are determined based on the spectator field identification method of the spectator terminal, wherein the spectator field identification method includes spectator terminal input and marking Assessment identification.
  • the recommended marking live broadcast determination module 430 is set to:
  • the recommended marking live broadcast of the spectator terminal is determined.
  • the live broadcast status includes the duration of the live broadcast of the current live broadcast object, and the number of the recommended marking live broadcast is at least one;
  • the recommended marking live broadcast determination module 430 is also set to:
  • the recommendation priority of at least one recommended marking live broadcast is determined based on the live broadcast status of at least one recommended marking live broadcast.
  • the device also includes:
  • the marking information evaluation module is configured to evaluate the marking information after receiving the marking information fed back by the spectator, and determine the evaluation result of the marking information
  • the spectator field updating module is configured to update the spectator field of the spectator terminal in the spectator field library based on the evaluation result of marking information in at least one field category.
  • the device also includes:
  • the marking setting module is configured to mark the at least one live data based on the marking information of at least one spectator terminal, and update the marking corresponding to the at least one live data to the marking library;
  • the video recommendation module is configured to receive the recall request sent by the client, based on the mark and/or user portrait in the recall request, and the mark of the at least one live data in the marking library, determine the recommended live data, and The recommended live broadcast data is sent to the client.
  • the live broadcast platform provided in the embodiment of the present application can execute the live broadcast processing method provided in any embodiment of the present application, and has corresponding functional modules for executing the live broadcast processing method.
  • FIG. 6 is a schematic structural diagram of a live broadcast platform provided by an embodiment of the present application.
  • the live broadcast platform includes: a login module 100 , a spectator domain database 200 , a spectator domain query module 300 , a live broadcast domain determination module 400 , a recommendation module 500 , and a marking evaluation module 600 .
  • the login module 100 is configured to make a login judgment when requested by the user. If not logged in, jump to the system login module to log in, and request again after the login is completed. After the user logs in, the authority is judged. There are two types of authority related to spectators: no spectator authority (the page prompts to register as a spectator), and spectator authority. When there is no spectator permission, the user is prompted to register as a spectator on the platform. After successful registration, you have spectator permission.
  • the spectator field library 200 includes a storage unit, which includes a positive data structure and a reverse data structure, wherein the positive data structure is a redis-set structure, the key is a domain classification id, the value is a set of spectator ids, and the reverse data structure is redis-zset structure, key is the viewer id, value is the collection of category ids, and score is the authoritative score value of the viewer in the category.
  • a forward index is established in the spectator field library 200, and its query logic is: through the category id, you can query the set of all spectators, and an inverted index is established. The query logic is: through the spectator id, you can query the category collection that it is good at, and you can view to the authority score value for each category.
  • the spectator domain library 200 can perform query operations, new operations and update operations.
  • the query operation uses the redis-zset.zrevrange method to query the top N categories, which requires 1 parameter: spectator id; the new operation uses redis.zset.zadd The method adds elements to the collection, which requires three parameters: spectator id, category id, and category score; the update operation is updated through the redis.zset.zadd method.
  • the spectator field query module 300 is set to match the field that the spectator is best at, and recommend after matching the field that the spectator is best at with the live broadcast room.
  • the viewer ID is obtained, that is, the viewer id.
  • the query result is 0, the query result is 1, and the query result is 2.
  • the initialization logic is executed: display the field classification information to the spectator, prompt the spectator to choose the 3 best, and receive the 3 selected by the spectator according to the user's judgment.
  • a field of expertise (category id), write the record into the spectator field database.
  • the key is the spectator id
  • the value is the category id
  • the score is written as 0.
  • the query result 1, it means that the spectator has only performed the initial entry, and the result is marked as low.
  • the query result 2 it means that the spectator has been intelligently identified the authority field, and the result is marked as high.
  • the live broadcast area determination module 400 is configured to obtain the theme styles of all the live broadcast rooms being broadcast, and recommend the live broadcast to viewers.
  • the live broadcast product is obtained based on the live broadcast platform, so as to obtain the category of the product.
  • the redis-set structure for data storage the key is the category id, and the value is the live id.
  • the server will be called back to remove the live id from the set.
  • One of the live broadcasts may have multiple products, but the main category id is generally the same.
  • the recommendation module 500 is configured to map and match the live room and the viewer end, and recommend a suitable live room for the viewer end.
  • FIG. 7 is a schematic diagram of the live room recommendation provided by the embodiment of the present application.
  • the query result of the spectator field query module 300 is 2, that is, when the result is marked as high, the categories that the spectator terminal is good at are a1, a2, and a3 through intelligent identification.
  • Select 4 categories, 3 of which are the categories that the spectator is good at, and 1 is a random category that does not include the first three categories, in order to achieve a balance between applying the optimal ability of the spectator and exploring the potential ability. Randomly select one of the above 4 categories, and push the live room id generated by the field classification identifier, that is, the classification id query, to the viewer side, and the viewer side sees a new live broadcast when the screen slides down.
  • the query result of the spectator field query module 300 is 1, that is, when the result is marked as low, the self-selected category of the spectator terminal without intelligent identification is a1, a2.
  • Select 4 categories, 2 of which are selected by the spectators as set A, and 2 random categories that do not include the first two categories are recorded as set B. Randomly select a category id from set A and set B for each query, use the above two category ids to query a live room id respectively, and push the live room id to the spectator side, and the performance seen on the spectator side
  • the marking evaluation module 600 is set to automatically update the viewer's authoritative field to better identify the viewer, thereby recommending more matching live broadcast rooms to the viewer.
  • the spectator side watches, diagnoses, and saves the diagnosis results in the live broadcast room.
  • the spectator diagnosis is successful, it obtains the category id of the successful spectator diagnosis, and updates the spectator domain database 200, where the key is the spectator id, the value is the category id, and the score is the original value +1, if there is no record, the score defaults to 0. Every time the viewer diagnoses the results, the corresponding domain database will be updated, so as to recommend a more suitable live room next time, so as to realize the automatic upgrade of the recommendation ability.
  • FIG. 8 is a schematic structural diagram of a live broadcast processing device provided in Embodiment 5 of the present application, and the live broadcast processing device may be configured at a spectator end.
  • the unit includes:
  • the first data request module 510 is configured to send a recall request to the live broadcast platform, and receive the live broadcast identification of the recommended marking live broadcast that matches the spectator field of the current device;
  • the second data request module 520 is configured to generate a live data request based on the selected live broadcast identification of the recommended marking live broadcast, send the live data request to the live broadcast platform, and receive the live data fed back by the live broadcast platform and the live data to be printed. label information;
  • the marking module 530 is configured to display the live data and the information to be marked, and obtain the marking information corresponding to the information to be marked;
  • the marking information feedback module 540 is configured to send the marking information and the live identification corresponding to the live data to the live broadcast platform, so that the live broadcast platform stores the marking information of the live data, wherein the The marking information is recommended reference information corresponding to the live data in the live broadcast platform.
  • the device also includes:
  • the field classification information display module is configured to receive the field classification information fed back by the live broadcast platform, and display the field classification information on the display interface;
  • the field classification information feedback module is configured to detect the selection information of the field classification information input by the user, and send the selected field classification information to the live broadcast platform, wherein the selected field classification information is used to determine the audience field.
  • the marking information feedback module 540 is set to:
  • the information to be marked is displayed.
  • the live broadcast processing device provided in the embodiment of the present application can execute the live broadcast processing method provided in any embodiment of the present application, and has corresponding functional modules for executing the live broadcast processing method.
  • FIG. 9 is a schematic structural diagram of a live broadcast system provided in Embodiment 6 of the present application.
  • the live broadcast system includes:
  • the live broadcast terminal 610 is configured to acquire at least one live broadcast data and send at least one live broadcast data to the live broadcast platform;
  • the spectator terminal 620 is configured to send a recall request to the live broadcast platform
  • the live broadcast platform 630 is configured to determine the spectator field of the spectator end based on the spectator identification of the spectator end, obtain the current live field of the at least one live data, and determine the recommended marking live broadcast of the spectator end based on the spectator field and the current live field of the at least one live data , and send the recommended marking live broadcast to the audience;
  • the viewer terminal 620 receives the recommended marking live broadcast, displays the recommended marking live broadcast, collects the marking information of the recommended marking live broadcast, and sends the marking information to the live broadcast platform;
  • the live broadcast platform 630 is also configured to receive the marking information fed back by the spectators, and store the marking information of the at least one live data, wherein the marking information is recommended reference information corresponding to the live data.
  • the live broadcast system also includes a client, the client is connected to the live broadcast platform in communication, and the client is configured to send a recall request to the live broadcast platform;
  • the live broadcast platform is also configured to receive marking information fed back by at least one spectator terminal, set a mark for the at least one live data based on the marking information of the at least one spectator end, and update the mark corresponding to the at least one live data to the mark Library;
  • receiving the recall request sent by the client based on the mark and/or user portrait in the recall request, and the mark of the at least one live data in the marking library, determine the recommended live data, and send the recommended live data Data is sent to the client.
  • the live broadcast of the recommended marking live broadcast by determining the recommended marking live broadcast of the spectator end based on the audience field and the current live broadcast field of the at least one live broadcast data in at least one current live broadcast data, the live broadcast of the recommended marking live broadcast The identification is sent to the spectator end, so that the spectator end marks the recommended marking live broadcast. Since the recommended marking live broadcast is determined according to the field of viewers, within the field of expertise of the audience, it is beneficial for the audience to mark the live data in the field they are good at, and improve the accuracy and efficiency of marking information. At the same time, the marking information of at least one live data is stored, so that when the client requests live data, accurate recommendations can be made to the client based on the marking information, and the accuracy of live data recommendation can be improved.
  • FIG. 10 is a schematic structural diagram of an electronic device provided in Embodiment 7 of the present application.
  • FIG. 10 shows a block diagram of an electronic device 12 suitable for implementing embodiments of the present application.
  • the electronic device 12 shown in FIG. 10 is only an example, and should not limit the functions and scope of use of the embodiment of the present application.
  • Device 12 is typically an electronic device undertaking image classification functions.
  • electronic device 12 takes the form of a general-purpose computing device.
  • Components of the electronic device 12 may include, but are not limited to: at least one processor 16, a storage device 28, and a bus 18 connecting various system components (including the storage device 28 and the processor 16).
  • Bus 18 represents at least one of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus structures.
  • these architectures include but are not limited to Industry Standard Architecture (Industry Standard Architecture, ISA) bus, Micro Channel Architecture (Micro Channel Architecture, MCA) bus, Enhanced ISA bus, Video Electronics Standards Association (Video Electronics Standards Association, VESA) local bus and peripheral component interconnect (Peripheral Component Interconnect, PCI) bus.
  • Electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 12 and include both volatile and nonvolatile media, removable and non-removable media.
  • Storage device 28 may include computer system readable media in the form of volatile memory, such as random access memory (Random Access Memory, RAM) 30 and/or cache memory 32 .
  • Electronic device 12 may include other removable/non-removable, volatile/nonvolatile computer system storage media.
  • storage system 34 may be configured to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 10, commonly referred to as a "hard drive").
  • a disk drive for reading and writing to a removable non-volatile disk may be provided, as well as a removable non-volatile disk (such as a Compact Disc- Read Only Memory, CD-ROM), Digital Video Disc (Digital Video Disc-Read Only Memory, DVD-ROM) or other optical media) CD-ROM drive.
  • each drive may be connected to bus 18 via at least one data medium interface.
  • the storage device 28 may include at least one program product having a set (for example, at least one) of program modules configured to perform the functions of the various embodiments of the present application.
  • Program modules 26 generally perform functions and/or methods in the embodiments described herein.
  • the electronic device 12 can also communicate with at least one external device 14 (such as a keyboard, pointing device, camera, display 24, etc.), and can also communicate with at least one device that enables the user to interact with the electronic device 12, and/or communicate with the device that enables the user to interact with the electronic device 12.
  • Electronic device 12 is capable of communicating with any device (eg, network card, modem, etc.) that communicates with at least one other computing device. Such communication may occur through input/output (I/O) interface 22 .
  • the electronic device 12 can also communicate with at least one gateway (such as a local area network (Local Area Network, LAN), wide area network, Wide Area Network, WAN) and/or a public gateway, such as the Internet, through the gateway adapter 20.
  • LAN Local Area Network
  • WAN Wide Area Network
  • public gateway such as the Internet
  • gateway adapter 20 communicates with other modules of electronic device 12 via bus 18 .
  • other hardware and/or software modules may be used in conjunction with electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, disk arrays (Redundant Arrays) of Independent Disks, RAID) systems, tape drives, and data backup storage systems.
  • the processor 16 executes various functional applications and data processing by running the programs stored in the storage device 28 , for example, realizing the live broadcast processing method provided by the above-mentioned embodiments of the present application.
  • Embodiment 8 of the present application provides a computer-readable storage medium, on which a computer program is stored.
  • the program is executed by a processor, the live broadcast processing method provided in the embodiment of the present application is implemented.
  • the computer-readable storage medium provided by the embodiment of the present application is not limited to the method operation described above, and can also execute the live broadcast processing method provided by any embodiment of the present application.
  • the computer storage medium in the embodiments of the present application may use any combination of at least one computer-readable medium.
  • the computer readable medium may be a computer readable signal medium or a computer readable storage medium.
  • a computer readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof.
  • a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.
  • a computer readable signal medium may include a data signal carrying computer readable source code in baseband or as part of a carrier wave traveling as a data signal. Such propagated data signals may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • a computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. .
  • Source code embodied on a computer readable medium may be transmitted using any appropriate medium, including - but not limited to wireless, wire, optical cable, radio frequency (RF), etc., or any suitable combination of the foregoing.
  • any appropriate medium including - but not limited to wireless, wire, optical cable, radio frequency (RF), etc., or any suitable combination of the foregoing.
  • Computer source code for carrying out the operations of this application may be written in one or more programming languages, or combinations thereof, including object-oriented programming languages—such as Java, Smalltalk, C++, and conventional Procedural Programming Language - such as "C" or a similar programming language.
  • the Source 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.
  • the remote computer may be connected to the user computer through any kind of gateway, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (such as through an Internet Service Provider). Internet connection).
  • LAN local area network
  • WAN wide area network
  • Internet Service Provider such as AT&T, MCI, Sprint, EarthLink, MSN, GTE, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

本申请实施例公开了直播处理方法、直播平台、装置、系统、介质及设备,其中方法包括:获取看客端的召回请求,其中,看客端设置为对推荐直播进行打标;获取至少一个直播数据的当前直播领域,基于所述看客端的看客领域以及所述至少一个直播数据的当前直播领域确定看客端的推荐打标直播;将推荐打标直播的直播标识发送至看客端,并接收看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,打标信息为对应直播数据的推荐参考信息。

Description

直播处理方法、直播平台、装置、系统、介质及设备
本申请要求在2021年5月28日提交中国专利局、申请号为202110594732.8的中国专利申请的优先权,该申请的全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及计算机技术领域,例如涉及直播处理方法、直播平台、装置、系统、介质及设备。
背景技术
直播平台上一般有三种主要参与方:直播者、观看者和变现渠道,其核心链路是直播者汇集观看者后进行变现。常见的例子是各种平台提供直播服务,直播者对产品进行展示,观众观看直播,提升观众对品牌的认知或促成下单转化。
相关技术中至少存在以下技术问题:直播平台上包括大量的直播,且每一场直播中一般会播十几个甚至几十上百个品,每个品几分钟不等,绝大多数直播的时间有很大随机性。用户如果不全程观看,就容易错过想看的商品展示;用户如果全程观看,会浪费大量时间在并不关注的商品上。
发明内容
本申请实施例提供直播处理方法、直播平台、装置、系统、介质及设备,以实现向看客端进行精准推荐,以提高推荐打标视频与看客端的匹配度,提高打标信息的准确度。
第一方面,本申请实施例提供了一种直播处理方法,包括:
获取看客端的召回请求,其中,所述看客端设置为对推荐直播进行打标;
获取至少一个直播数据的当前直播领域,基于所述看客端的看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播;
将所述推荐打标直播的直播标识发送至所述看客端,并接收所述看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,所述打标信息为对应直播数据的推荐参考信息。
第二方面,本申请实施例还提供了一种直播处理方法,包括:
向直播平台发送召回请求,并接收与当前设备的看客领域相匹配的推荐打标直播的直播标识;
基于选择的推荐打标直播的直播标识生成直播数据请求,将所述直播数据请求发送至所述直播平台,并接收所述直播平台反馈的直播数据以及待打标信息;
显示所述直播数据以及所述待打标信息,并获取所述待打标信息对应的打标信息;
将所述打标信息以及所述直播数据对应的直播标识发送至所述直播平台,以使所述直播平台存储所述直播数据的打标信息,其中,所述打标信息为所述直播平台中对应所述直播数据的推荐参考信息。
第三方面,本申请实施例还提供了一种直播平台,包括:
召回请求获取模块,设置为获取看客端的召回请求,其中,所述看客端设置为对推荐直播进行打标;
当前直播领域获取模块,设置为获取至少一个直播数据的当前直播领域;
推荐打标直播确定模块,设置为基于所述看客端的看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播;
打标信息获取模块,设置为将所述推荐打标直播的直播标识发送至所述看客端,并接收所述看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,所述打标信息为对应直播数据的推荐参考信息。
第四方面,本申请实施例还提供了一种直播处理装置,包括:
第一数据请求模块,设置为向直播平台发送召回请求,并接收与当前设备的看客领域相匹配的推荐打标直播的直播标识;
第二数据请求模块,设置为基于选择的推荐打标直播的直播标识生成直播数据请求,将所述直播数据请求发送至所述直播平台,并接收所述直播平台反馈的直播数据以及待打标信息;
打标模块,设置为显示所述直播数据以及所述待打标信息,并获取所述待打标信息对应的打标信息;
打标信息反馈模块,设置为将所述打标信息以及所述直播数据对应的直播标识发送至所述直播平台,以使所述直播平台存储所述直播数据的打标信息,其中,所述打标信息为所述直播平台中对应所述直播数据的推荐参考信息。
第五方面,本申请实施例还提供了一种直播系统,包括:
直播端,设置为获取至少一个直播数据,并将所述至少一个直播数据发送至直播平台;
看客端,设置为向直播平台发送召回请求;
所述直播平台,设置为获取所述至少一个直播数据的当前直播领域,基于所述看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播,将所述推荐打标直播的直播标识发送至所述看客端;
所述看客端对所述推荐打标直播,展示所述推荐打标直播,并采集对所述推荐打标直播的打标信息,将所述打标信息发送至所述直播平台;
所述直播平台还设置为接收所述看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,所述打标信息为对应直播数据的推荐参考信息。
第六方面,本申请实施例还提供了一种电子设备,包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行所述程序时实现如本申请任一实施例提供的直播处理方法。
第七方面,本申请实施例还提供了一种计算机可读存储介质,其上存储有计算机程序,该程序被处理器执行时实现如本申请任一实施例提供的直播处理方法。
附图说明
图1为本申请实施例一提供的一种直播处理方法的流程示意图;
图2是本申请实施例提供一种应用场景的示意图;
图3为本申请实施例二提供的一种直播处理方法的流程示意图;
图4是本申请实施例三提供的一种直播处理方法的流程示意图;
图5是本申请实施例四提供的一种直播平台的结构示意图;
图6是本申请实施例提供的一种直播平台的结构示意图;
图7是本申请实施例提供的直播间推荐的示意图;
图8是本申请实施例五提供的一种直播处理装置的结构示意图;
图9是本申请实施例六提供的一种直播系统的结构示意图;
图10为本申请实施例七提供的一种电子设备的结构示意图。
具体实施方式
下面结合附图和实施例对本申请作详细说明。
实施例一
图1为本申请实施例一提供的一种直播处理方法的流程示意图,本实施例可适用于在大量直播数据中向对直播数据进行打标的看客端进行精准推荐的情况,该方法可以由本申请实施例提供的一种直播平台来执行,该直播平台可以由软件和/或硬件来实现,该直播平台可以配置在电子计算设备上,包括如下步骤:
S110、获取看客端的召回请求,其中,所述看客端设置为对推荐直播进行打标。
S120、获取至少一个直播数据的当前直播领域,基于所述看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播。
S130、将所述推荐打标直播的直播标识发送至所述看客端,并接收所述看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,所述打 标信息为对应直播数据的推荐参考信息。
示例性的,参见图2,图2是本申请实施例提供一种应用场景的示意图。直播端、看客端和客户端分别与直播平台通信连接,在一些实施例中,直播端、看客端和用户端可以是分别的三端,即三个不同的电子设备;在另一些实施例中,还可以是直播端、看客端和用户端中的任意两项或三项是同一电子设备,通过不同的展示界面实现不同用户权限的区分,或者通过登录不同的应用程序(Application,APP)进行区分,或者通过验证不同的角色身份进行区域。直播端、看客端和客户端可以是诸如手机、平板电脑等的终端设备,分别具有不同的权限,直播端设置为提供直播数据,具有直播后台管理权限、直播对象设置权项和直播效果报告查询权限。看客端可观看直播,并对直播进行评价,具有直播评价权限、平台评价权限,客户端可观看直播,仅具有直播观看权限。其中,看客端和客户端均具有对直播中直播对象的交易权限。
直播端、看客端分别需要进行注册,直播平台配置有注册模块,接收用户的注册请求以及用户信息,对用户的直播身份或看客身份进行注册,注册成功后具备相应的权限。对于任一设备向直播平台发起请求时,判断该设备是否登录,登录成功后,确定该设备的权限,即确定该设备为直播端、看客端和客户端中的一项。对于直播端可开设直播,设置直播对象(例如进行介绍的物品或者游戏等),生成直播数据,将生成的直播数据实时传输至直播平台,便于看客端和客户端从直播平台获取直播数据。本实施例中,通过设置可进行直播观看的看客端,增加了真实流量的比例,相应的减少了虚拟流量的比例,减少了恶意刷流量的情况。
基于看客端的数量大、用户擅长领域不同特性以及直播数据包括领域范围广的特征,向看客端推荐看客端用户不擅长领域的直播数据进行打标时,存在打标正确率低、打标效率差等的问题,影响打标信息的采集效率以及打标信息的正确性。本实施例中,在获取看客端的召回请求时,确定看客端的看客领域,该看客端的看客领域为看客端用户的擅长领域。其中,看客端的召回请求可以 是在看客端登录成功时触发,或者在看客端检测到刷新操作时触发。看客端的召回请求中包括看客端的看客标识,例如可以是看客端ID,或者看客端操作用户的标识等。
可选的,在获取看客端的召回请求后,以及基于所述看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播之前,还包括基于召回请求中看客标识确定看客端的看客领域。示例性的,基于召回请求中看客标识确定看客端的看客领域,包括:将所述看客端的看客标识在预设的看客领域库中进行匹配,确定所述看客标识对应的至少一个看客领域。其中,看客领域库中预先设置有领域分类标识,即不同的直播领域对应不同的领域分类标识,该领域分类标识可根据直播领域的变更进行更新,例如,新增领域分类标识或者修改领域分类标识等。看客领域库中存储有看客标识以及看客标识对应的领域分类标识。在一些可选实施例中,看客领域库好存储有看客标识-领域标识的存储数据结构,该存储数据结构可以是redis-set结构,key为领域分类标识id,value为看客标识(例如看客id)集合,该存储数据结构还可以是redis-zset结构,key为看客标识(例如看客id),value为领域分类标识id集合,score为看客在该分类的权威评分值。通过不同结构进行数据存储,可基于不同索引信息进行快速查询。示例性的,在redis-set结构的存储数据中,可通过领域分类标识id查询该领域下的看客标识集合,在redis-zset结构的存储数据中,可通过看客标识查询对应看客所擅长的领域分类标识集合。
看客领域库中每个看客端对应的领域分类标识可以是通过两种方式确定,其一是在看客端首次登录直播平台时,直播平台向看客端发送领域分类信息,看客端采集用户选择的预设数量(例如可以是两个)的领域分类信息,将选择的领域分类信息的领域分类标识发送至直播平台,直播平台基于接收到的看客标识以及领域分类标识更新看客领域库,且看客端对领域分类标识的权威评分值score记为零。其二,监测看客端进行打标的历史直播数据,统计历史直播数据的领域分类信息以及对历史直播数据进行打标的打标评估,确定看客端的擅 长领域信息,基于擅长领域信息和看客标识更新看客领域库。看客领域库中不同看客标识对应的领域分类标识的数量可以不同,例如,可以是两个或三个等,对此不做限定。在一些可选实施例中,可以是基于看客端对应的领域分类标识的确定方式,对看客标识设置识别标识,例如,看客端对应的领域分类标识的确定方式包括看客端输入和打标评估识别,看客端输入的识别标识可以是1,打标评估识别的识别标识可以是2。
本实施例中,可以是在接收到看客端的召回请求时,基于召回请求中的看客标识在redis-set结构的存储数据中进行看客标识的匹配,得到该看客标识对应的领域分类标识集合,基于该集合中的领域分类标识确定看客领域,相应的,在看客领域库查询的结果可以是1或2。看客标识在看客领域库中匹配失败,即在看客领域库查询的结果可以是0,表明看客领域库中未存储该看客标识对应的领域分类标识,向看客端发送领域分类信息,接收所述看客端反馈的看客领域,并更新所述看客领域库。
在确定看客领域后,可将与看客领域相匹配的直播数据作为对应看客端的推荐打标直播。相应的,直播数据的直播领域可以是唯一的,可以是根据直播对象的变化而变化。
对于任一直播数据,基于所述直播数据中的核心直播对象确定所述直播数据的当前直播领域,其中,核心直播对象可以是直播端设置的,例如可以是直播时长大于直播总时长一半的直播对象。任一直播对象(例如商品、货物或者网络游戏等)均配置有至少一级类目信息,可以是基于最小级别的类目信息确定直播对象的领域分类标识。示例性的,面巾纸的类目信息包括生活用品/纸品/面巾纸,相应的,上述类目信息的级别依次下降,根据最小级别的类目信息即面巾纸确定领域分类标识,例如面巾纸对应的领域分类标识可以是Z-01,对领域分类标识的形式不作限定。
在一些实施例中,获取每个直播数据的当前直播领域,包括:基于所述直播数据中的当前直播对象确定所述直播数据的当前直播领域。在同一次直播中 可以是包括多个直播对象,且不限定直播对象的类型,因此在同一直播数据中,不同时间段可以对应不同的直播对象,相应的,直播数据对应的直播领域同样发生变化,例如前一直播对象为手机,下一直播对象为速冻水饺,相应的,直播领域随直播对象发生变化。检测直播数据中直播对象的切换,在检测到直播对象发生切换时,基于切换后的直播对象更新直播数据的当前直播领域,例如基于切换后的直播对象的最小级别的类目信息确定直播数据的当前直播领域。可选的,可以是通过识别直播数据中的语音关键词监测直播对象的切换,可以是通过对直播数据中图像进行图像识别,确定当前展示的直播对象实现对直播对象的切换监测,还可以是通过直播数据中对象接口的监测实现对直播对象的切换监测,其中,对象接口为访问直播对象的接口,例如详情页访问接口、购买接口、直播页面的当前直播商品接口等。
获取至少一个直播数据的当前直播领域,形成直播数据的领域数据库,该直播数据的领域数据库可以实时变化的,监测到新增直播数据、直播结束或直播数据的当前直播领域发生变化时,更新直播数据的领域数据库。将看客领域在直播数据的领域数据库中进行匹配,以确定每个看客端的推荐打标直播,示例性的,将看客端的至少一个看客领域分别在直播数据的领域数据库中进行匹配,确定至少一个看客领域对应的直播数据,在上述至少一个看客领域对应的直播数据中确定推荐打标直播。通过基于看客领域确定推荐打标直播,该推荐打标直播与看客端用户所擅长的领域相匹配,有利于看客端快速准确的对推荐打标直播进行打标,提高向看客端推荐直播数据的针对性和准确性,提高直播数据所获取的打标信息的效率和准确性,相应的,在基于打标信息向客户端进行直播数据推荐时,基于高精度的打标信息,可提高直播数据推荐的针对性和精确度。
在一些可选实施例中,看客端的推荐打标直播包括:所述看客领域对应的第一直播和所述看客领域以外的其他领域分类的第二直播,所述第一直播和所述第二直播的数量满足预设比例。其中,第一直播可以是基于看客端的看客领 域与直播数据的当前直播领域匹配确定的,第二直播可以是在看客领域以外的其他领域分类中随机确定的直播领域,以及在该随机确定的直播领域对应的直播数据中随机确定的。通过向看客端推荐看客端非擅长的领域的直播数据,有利于发掘看客端的擅长领域以及优化看客领域库。
第一直播和第二直播的数量满足预设比例,该预设比例可以是在1:1-5:1之间,可根据用户需求调节预设比例。
在一些可选实施例中,第一直播和所述第二直播满足的预设比例基于所述看客端的看客领域识别方式确定,其中,所述看客领域识别方式包括看客端输入和打标评估识别。示例性的,若看客领域识别方式为看客端输入,则第一直播和所述第二直播的预设比例可以是1:1,若看客领域识别方式为打标评估识别,则第一直播和所述第二直播的预设比例可以是3:1。在看客领域识别方式为看客端输入时,看客领域未经过验证,增大第二直播的比例,便于快速确定适用于该看客端的领域分类标识。
将推荐打标直播的直播标识(例如直播间ID)发送至看客端,以使看客端根据推荐打标直播选择进行打标的直播数据,基于选择的直播标识向直播平台发送访问请求。直播平台接收看客端基于推荐打标直播发送的访问请求,向看客端反馈直播数据和对应的待打标信息。在一些可选实施例中,将所述访问请求对应的直播数据以及所述直播数据对应的待打标信息发送至所述看客端,包括:将所述访问请求对应的直播数据发送至所述看客端,并进行直播显示计时;若所述直播显示计时满足预设显示时间,则将所述直播数据对应的待打标信息发送至所述看客端。通过设置直播显示计时,保证看客端基于直播显示了解直播内容,提高打标信息的准确性,避免在未观看直播的情况下进行随机打标的情况。在一些可选实施例中,将所述访问请求对应的直播数据以及所述直播数据对应的待打标信息发送至所述看客端,包括:将直播数据和对应的待打标信息同步发送至看客端,以向看客端展示直播数据,并在直播显示时间满足预设显示时间时显示待打标信息。本实施例中,通过设置看客端,该看客端通过对 直播数据进行预设时长的展示以及打标后进行下一直播数据的展示和打标,可将同一看客端的流量服务于多个直播间,提供了一种新的直播观看模式,有利于打破互联网群体总数固定导致的流量瓶颈。
可选的,直播端还设置为设置部分待打标消息,并将该部分待打标消息发送至直播平台,直播平台将接收的直播端发送的部分待打标消息,以及预先设置的另一部分待打标消息,形成该直播端的直播数据对应的待打标消息。该待打标消息可以是与直播的标识(例如直播ID(即直播间ID)或直播端ID)进行关联。示例性的,待打标消息可以包括但不限于食物是否诱人、主播试吃效果是否满意、主播吃完后花妆感是否明显等,另一部分待打标消息可以是当前直播是否涉嫌违规、当前直播间是否在挂机(主播不在摄像头前)、当前直播售卖品类是什么(衣服/食品)。通过由直播端和直播平台共同确定待打标消息,便于确定每个直播的直播状态,避免恶意直播或非法直播等。待打标消息可以是根据直播类型以及直播对象设置,对此不作限定。上述直播标识对应的待打标消息可以是存储在打标库中,便于根据直播标识调用对应的待打标消息。
在上述实施例的基础上,向看客端发送的待打标信息中包括已知标记的待打标信息和未知标记的待打标信息;其中,已知标记的待打标信息可以是具有固定标记的待打标信息,该待打标信息的已知标记可以是通过对直播画面进行识别得到,示例性的,已知标记的待打标信息可以是但不限于画面中的直播人数,主播是否在画面中等。
在获取至少一个看客端反馈的打标信息之后,以及基于至少一个所述看客端的打标信息对所述直播数据设置标记之前,还包括:基于所述已知标记对所述至少一个看客端反馈的打标信息进行验证,即对于已知标记的待打标信息,将已知标记和看客端反馈的打标信息进行比对,若二者一致,则确定验证成功,若二者不一致,则确定验证失败。验证失败时,表明看客端为非正常的打标状态,例如机器打标或者随机打标等,丢弃所述看客端反馈的打标信息,避免错误打标信息对正确打标的干扰。对于验证成功的标记信息,统计至少一个看客 端反馈的打标信息,并确定所述打标信息中的目标打标信息,基于所述目标打标信息验证所述至少一个看客端反馈的打标信息的正确性。其中,对于任一待打标信息,存在预设数量以及预设比例的相同打标信息,即相同打标信息数量满足预设数量,且该预设数量满足占整体打标信息的预设比例,则将该相同的打标信息确定为目标打标信息,其中,预设比例可以是70%,预设数量可以是500或1000等,可根据需求调节。
可选的,在任一目标打标信息属于预警打标信息时,生成预警提示信息,并发送预警提示信息至直播端,或者屏蔽对应的直播数据。
基于目标打标信息验证所述至少一个看客端反馈的打标信息的正确性,即对于同一待打标信息,与目标打标信息不同的打标信息确定为不正确。对正确的打标信息形成一打标日志,用于记录看客端标识、直播标识、打标时间戳、待打标信息以及打标信息。本实施例中,通过存储至少一个直播数据的打标消息,将打标消息作为对应视频数据的推荐参考信息,提高了对至少一个直播数据的推荐针对性。
在上述实施例的基础上,在接收所述看客端反馈的打标信息之后,所述方法还包括:对所述打标信息进行评估,确定所述打标信息的评估结果;基于至少一个领域分类中打标信息的评估结果更新所述看客领域库中看客端的看客领域。其中,对打标信息进行评估,可以是确定打标信息的正确率以及打标效率,即评估结果包括但不限于打标正确率以及打标效率。对于任一领域分类,基于该领域分类对应的历史打标信息的评估结果以及打标数量确定该领域分类的擅长程度,基于不同领域分类的擅长程度的排序,确定看客端当前的看客领域,并更新看客领域库。
在一些实施例中,可以是基于预设时间间隔更新看客领域库,例如预设时间间隔可以是一天、两天或者一星期等。
本实施例的技术方案,通过在当前的至少一个直播数据中,基于看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播, 将所述推荐打标直播的直播标识发送至所述看客端,以使看客端对推荐打标直播进行打标。由于推荐打标直播是根据看客领域针对性确定的,在看客端的擅长领域范围内,有利于看客端针对所擅长领域内的直播数据进行打标,提高打标信息的正确率和打标效率。同时将至少一个直播数据的打标信息进行存储,便于在客户端请求直播数据时,基于打标信息向客户端进行精准推荐,提高直播数据的推荐精准度。
在上述实施例的基础上,所述方法还包括:基于至少一个所述看客端的打标信息对所述至少一个直播数据设置标记,并将至少一个直播数据对应的标记更新至打标库;接收客户端的发送的召回请求,基于所述召回请求中的标记和/或用户画像,以及所述打标库中至少一个直播数据的标记,确定推荐直播数据,将所述推荐直播数据发送至所述客户端。
对于任一直播,将确定至少一个目标打标信息形成该直播标识对应的标记,例如待打标信息为主播语速是否过快,目标打标信息为是,则确定标记为语速过快。将每个直播标识对应的至少一个标记,更新存储至打标库中,该打标库在通过redis对每个直播的标记进行保存。
在接收到用户端发送的召回请求,该召回请求可以是通过平台应用启动触发,或者通过在平台应用中的刷新操作触发,或者通过在平台应用中的搜索操作触发。该召回请求中可以包括用于搜索的标记和用户画像的至少一项。其中,直播平台中可以是预先设置了用户画像与标记之间的对应关系,可基于用户画像确定对应的标记。将召回请求中的标记或通过用户画像确定的标记在打标库中进行标记匹配,确定标记对应的至少一个直播标识,将该直播标识对应的直播数据确定为推荐直播数据,推荐给客户端。在一些实施例中,看客端可作为用户端。
通过看客端进行打标形成的标记,对直播数据进行真实打标,通过该标记为客户端匹配对应的直播数据,提高了匹配的准确度以及直播的推荐精度,有利于提高直播转化率。
本实施例的技术方案,通过增加看客端将直播数据与观看用户紧密结合,避免了直播体验割裂的问题。对直播数据进行实时在线打标,实现直播内容实时感知,便于直播平台实时获取至少一个直播的真实状态以及对至少一个直播通过标记进行准确定位,在向客户端进行推荐时,根据打标库中的标记匹配高准确度的推荐直播数据,避免向用户展示大量不关注的直播数据的情况。
在上述实施例的基础上,将至少一个直播数据对应的标记更新至打标库之后,所述方法还包括:基于至少一个直播数据、至少一个直播数据的标记以及标记对应的看客端生成打标报告,将所述打标报告发送至对应的直播端,其中,所述打标报告包括直播打标结果、看客打标结果、直播-标记关系以及标记看客贡献的至少一项。
其中,直播打标结果用于通过直播id查询当前直播的诊断结果,为redis-zset结构的数据缓存,key为直播id,value为打标集合,score为打标时间。如直播id-123456,打标集合:{value:“语速过快”,socre:1650151159、value:“灯光晃眼”,score:1590123123}。看客打标结果用于通过看客id查询看客的打标结果,为redis-zset结构的数据缓存。key为看客id,hash-key为直播id,hash-value为打标结果。如看客端a,打标直播id-123456,打标集合:{“语速过快”、“灯光正常”}。直播-标记关系用于通过标记查询直播列表,为redis-zset结构的数据缓存。key为标记,value为直播id,score为时间戳。如标记-衣物,value-{id:123,score:1505454122}。标记看客贡献用于在直播结束后统计看客端贡献度。为redis-set-list结构的数据缓存,key为标记,set-key为直播id,list为看客id列表。如标记-羊毛大衣,set-key-{id:1,list:[看客id:1,看客id:2]}。
通过形成打标报告,并将打标报告发送至直播端,以使直播用户可查询标记,以通过标记进行改善,有利于提高直播质量。
本实施例中,通过实时更新的打标库维持一个热的标记-直播缓存数据,用于存储当前预设时间间隔内至少一个标记与对应的至少一个直播的对应关系,其中,一个标记可以是对应至少一个直播标识,任一直播标识可以是对应有至 少一个标记。其中当前预设时间间隔可以是在当前时刻之前的预设时间间隔,例如5分钟。在标记-直播缓存数据中标记可以是基于更新的时间戳进行排序。需要说明的是,标记-直播缓存数据可以是实时更新的,还可以是基于固定时间间隔进行更新的。
可选的,基于所述打标库中的至少一个标记、至少一个标记的时间戳以及至少一个标记对应的直播数据,设置一预设时间间隔内的标记-直播缓存数据,包括:基于所述打标库中的至少一个标记、至少一个标记的时间戳以及至少一个标记对应的直播数据,获取当前预设时间间隔内的标记-直播缓存数据;创建数据更新定时任务,其中,所述数据更新定时任务用于在所述标记-直播缓存数据中增加新增时长中标记与直播数据的对应关系,并剔除所述标记-直播缓存数据中过期时长中标记与直播数据的对应关系;基于新增标记,执行所述数据更新定时任务对所述标记-直播缓存数据进行更新。
其中,新增时长可以是定时任务的时间间隔,例如10s,新增时长中标记可以是相邻定时任务之间的增量标记,即新增10s内接收的标记,将该标记与设置有该标记的直播数据建立对应关系,即将该标记与设置该标记的直播数据的直播标识设置对应关系。过期时长为预设时间间隔与当前时刻最远的固定时长,例如与当前时刻最远的10s,新增时长与过期时长相同。
随时间的变化,增加新增标记与直播数据的对应关系,剔除过期标记与直播数据的对应关系,维持固定时间间隔的标记-直播缓存数据。
在接收到召回请求时,确定进行查询的标记,将上述标记在标记-直播缓存数据中进行匹配,确定匹配成功的直播数据作为推荐直播数据。其中,由于标记-直播缓存数据中存储最新的标记,即反应了观看用户的最新体验,相应的,推荐直播数据为基于最新标记匹配确定,符合大量观看用户的推荐,提高了推荐直播数据的推荐精度。
实施例二
图3为本申请实施例二提供的一种直播处理方法的流程示意图,本申请实施例与上述实施例中各个可选方案可以结合。在本申请实施例中,可选地,基于所述看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播,包括:根据所述至少一个直播数据的直播状态,确定至少一个可推荐直播数据;基于所述看客领域以及所述至少一个可推荐直播数据确定所述看客端的推荐打标直播。
如图3所示,本申请实施例的方法包括如下步骤:
S210、获取看客端的召回请求,其中,所述看客端设置为对推荐直播进行打标。
S220、获取至少一个直播数据的当前直播领域,根据所述至少一个直播数据的直播状态,确定至少一个可推荐直播数据。
S230、基于所述看客领域以及所述至少一个可推荐直播数据确定所述看客端的推荐打标直播。
S240、将所述推荐打标直播的直播标识发送至所述看客端,并接收所述看客端反馈的打标信息,存储至少一个直播数据的打标信息,其中,所述打标信息为对应直播数据的推荐参考信息。
本实施例中,直播数据的直播状态可以是根据直播时长或者打标信息的采集量确定。可选的,直播状态可以是包括打标中和完成打标。可以将直播状态为打标中的直播数据确定为可推荐直播数据。
确定推荐打标直播的目的在于通过看客端对直播数据的打标,便于快速确定直播数据的直播内容,以精准向客户端进行推荐。例如,对于任一直播数据切换直播对象时,该直播数据的直播状态为打标中;若该直播对象的直播时长超过预设时长或者针对该直播对象接收的打标信息的数量超过预设数量,将直播数据的直播状态切换为完成打标,其中,预设时长可以是固定的时长,例如5分钟,或者还可以是根据直播对象的直播总时长确定,该预设时长可以是直播对象的直播总时长的一半。示例性的,若针对任一直播对象的直播状态处于直 播尾声或已接收到超出指定数量的打标信息,则表明针对该直播对象的打标已完成,对该直播数据进行打标推荐存在浪费看客端资源的情况,可将该直播数据确定为可不推荐直播数据,并在该直播数据切换直播对象时,将该直播数据确定为可推荐直播数据。
本实施例中,通过直播视频中的直播状态,在可推荐直播数据中确定推荐打标直播,在保证推荐打标直播的领域针对性的基础上,避免了无效打标推荐以及看客端资源的浪费,提高了推荐打标直播的有效性和精准度。
在上述实施例的基础上,直播状态包括当前直播对象的直播时长,所述推荐打标直播的数量为至少一个,相应的,在确定推荐打标直播之后,基于至少一个所述推荐打标直播的直播状态确定至少一个推荐打标直播的推荐优先级,并基于推荐优先级将推荐打标直播的直播标识发送至所述看客端。其中,推荐打标直播中当前直播对象的直播时长越大,表明该推荐打标直播的打标剩余时间越小,打标需求越紧急,推荐打标直播的推荐优先级越高,便于客户端对打标需求紧急的直播数据进行优先打标。
实施例三
图4是本申请实施例三提供的一种直播处理方法的流程示意图,本实施例可适用于看客对直播平台推荐的打标直播进行打标的情况,该方法可以由本申请实施例提供的一种直播处理装置来执行,该直播处理装置可以由软件和/或硬件来实现,该直播处理装置可以配置在看客端,看客端可以是诸如手机、平板电脑或者个人计算机(Personal Computer,PC)端等的电子设备。该方法包括:
S310、向直播平台发送召回请求,并接收与当前设备的看客领域相匹配的推荐打标直播的直播标识。
S320、基于选择的推荐打标直播的直播标识生成直播数据请求,将所述直播数据请求发送至所述直播平台,并接收所述直播平台反馈的直播数据以及待打标信息。
S330、显示所述直播数据以及所述待打标信息,并获取所述待打标信息对应的打标信息。
S340、将所述打标信息以及所述直播数据对应的直播标识发送至所述直播平台,以使所述直播平台存储所述直播数据的打标信息,其中,所述打标信息为所述直播平台中对应所述直播数据的推荐参考信息。
看客端在看客角色登录成功后,或者检测到用户输入的刷新操作,或者根据预设的刷新时间间隔生成召回请求,将该召回请求发送至直播平台。该召回请求中包括看客端的看客标识,看客标识可以是看客端的唯一标识或者看客端操作用户的唯一标识。直播平台可通过看客标识在看客领域库中进行匹配,以匹配到看客标识对应的看客领域,并根据看客领域确定相匹配的看客领域推荐打标直播,并将推荐打标直播的直播标识反馈至看客端。
在一些可选实施例中,在向直播平台发送召回请求之后,所述方法还包括:接收所述直播平台反馈的领域分类信息,并将所述领域分类信息在显示界面进行显示;检测用户输入的对领域分类信息的选择信息,将选择的领域分类信息发送至所述直播平台,其中,所述选择的领域分类信息用于确定看客领域。在直播平台基于看客标识未匹配到看客领域时,表明看客领域库中不包括该看客标识以及该看客标识对应的看客领域。将预先设置的领域分类信息发送至看客标识对应的看客端。看客端接收并显示领域分类信息,显示的领域分类信息可通过用户输入的选择操作区别显示,例如在任一领域分类被选择,可通过显示颜色或者显示尺寸的变化,或者在被选择的领域分类的前后设置选择标识进行区别显示,其中,选择标识可以是圆点或者星号等,对此不作限定。在一些实施例中,用户可选择预设数量的领域分类,例如可以是两个。可选的,领域分类信息的显示界面中可以是包括“确定”控件,在检测到“确定”控件被触发后,基于已选择的领域分类生成反馈信息,该反馈信息中包括已选择的领域分类的标识和看客标识,将该反馈信息发送至直播平台,以使该直播平台将反馈信息中的看客标识和领域分类标识更新至看客领域库中,并基于领域分类标识 确定看客领域。推荐打标直播与看客端用户所擅长的领域相匹配,有利于看客端快速准确的对推荐打标直播进行打标,提高向看客端推荐直播数据的针对性和准确性,提高直播数据所获取的打标信息的效率和准确性。
看客端接收推荐打标直播的直播标识,其中,该推荐打标直播包括:看客领域对应的第一直播和看客领域以外的其他领域分类的第二直播,第一直播和第二直播的数量满足预设比例。通过向看客端推荐看客端非擅长的领域的直播数据,有利于发掘看客端的擅长领域以及优化看客领域库。
看客端基于直播显示对推荐打标直播进行显示,其中,所述推荐打标直播的数量为至少一个,至少一个推荐打标直播的显示顺序可以是基于直播平台对至少一个推荐打标直播的推荐顺序确定。可选的,看客端显示的推荐打标直播可以是显示推荐打标直播的封面图像以及描述信息,其中,封面图像中可以是包括主播和/或直播对象,还可以是直播间预先设置的固定封面。描述信息可以是包括主播描述信息和/或直播对象的描述信息,便于用户通过显示内容获知推荐打标直播的相关信息。
看客端检测用户对显示的至少一个推荐打标直播的选择操作,示例性的,可以是检测对至少一个推荐打标直播的点击操作,例如触控点击操作或者通过鼠标或者键盘等输入的点击操作。基于点击操作对应的推荐打标直播的直播标识生成直播数据请求,发送至直播平台,该直播数据请求中包括直播标识和看客标识。
直播平台接收直播数据请求,基于直播标识匹配到对应的直播数据,以及对应的待打标信息,将匹配到的直播数据和待打标信息反馈至看客端。看客端在检测到对显示的任一推荐打标直播的选择操作后,进入直播显示界面,接收直播平台反馈的直播数据,在直播显示界面显示该直播数据。在一些可选实施例中,显示所述直播数据,并进行直播显示计时,若所述直播显示计时满足预设显示时长,则显示所述待打标信息。通过设置直播显示计时,保证看客端基于直播显示了解直播内容,提高打标信息的准确性,避免在未观看直播的情况 下进行随机打标的情况。
在一些实施例中,看客端可以是同步接收到直播数据和对应的待打标信息,还可以是先接收到直播数据,显示所述直播数据,并进行直播显示计时,若所述直播显示计时满足预设显示时长,则向直播平台发送触发信号,接收直播平台发送的待打标信息,并显示该待打标信息。
待打标信息中可以是包括多个问题,每一个问题可以是对应至少两个选项供用户选择,或者对应一信息输入位置,便于用户输入打标信息。看客端检测用于在显示界面对待打标信息的输入操作,例如对待打标信息的任一选项的选择操作,将选择的选项作为打标信息,或者在信息输入位置的输入操作,将输入的内容作为打标信息等。获取用户输入操作形成的打标信息,将每个待打标信息对应的打标信息反馈至直播平台,便于直播平台将至少一个看客端设置的打标信息作为推荐参考信息进行存储。
本实施例提供的技术方案,通过看客端在向直播平台发送召回请求时,接收直播平台反馈的与看客端的看客领域相匹配的推荐打标直播,可根据用户选择向直播平台请求选择的直播数据,并对该直播数据进行展示以及基于待打标信息对该直播数据进行打标,由于推荐打标直播与看客端用户所擅长的领域相匹配,有利于看客端快速准确的对推荐打标直播进行打标,提高向看客端推荐直播数据的针对性和准确性,提高直播数据所获取的打标信息的效率和准确性。
实施例四
图5是本申请实施例四提供的一种直播平台的结构示意图,该直播平台包括:
召回请求获取模块410,设置为获取看客端的召回请求,其中,所述看客端设置为对推荐直播进行打标;
当前直播领域获取模块420,设置为获取至少一个直播数据的当前直播领域;
推荐打标直播确定模块430,设置为基于所述看客领域以及所述至少一个直 播数据的当前直播领域确定所述看客端的推荐打标直播;
打标信息获取模块440,设置为将所述推荐打标直播的直播标识发送至所述看客端,并接收所述看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,所述打标信息为对应直播数据的推荐参考信息。
在上述实施例的基础上,召回请求获取模块410设置为:
将所述看客端的看客标识在预设的看客领域库中进行匹配,确定所述看客标识对应的至少一个看客领域;或者,
若所述看客领域库中不包括所述看客标识,则向看客端发送领域分类信息,接收所述看客端反馈的看客领域,并更新所述看客领域库。
在上述实施例的基础上,当前直播领域获取模块420设置为:
对于任一直播数据,基于所述任一直播数据中的核心直播对象确定所述任一直播数据的当前直播领域,或者,基于所述任一直播数据中的当前直播对象确定所述任一直播数据的当前直播领域。
在上述实施例的基础上,所述看客端的推荐打标直播包括:所述看客领域对应的第一直播和所述看客领域以外的其他领域分类的第二直播,所述第一直播和所述第二直播的数量满足预设比例。
在上述实施例的基础上,所述第一直播和所述第二直播满足的预设比例基于所述看客端的看客领域识别方式确定,其中,所述看客领域识别方式包括看客端输入和打标评估识别。
在上述实施例的基础上,推荐打标直播确定模块430设置为:
根据所述至少一个直播数据的直播状态,确定至少一个可推荐直播数据;
基于所述看客领域以及所述至少一个可推荐直播数据确定所述看客端的推荐打标直播。
可选的,所述直播状态包括当前直播对象的直播时长,所述推荐打标直播的数量为至少一个;
推荐打标直播确定模块430还设置为:
基于至少一个推荐打标直播的直播状态确定至少一个推荐打标直播的推荐优先级。
在上述实施例的基础上,该装置还包括:
打标信息评估模块,设置为在接收所述看客端反馈的打标信息之后,对所述打标信息进行评估,确定所述打标信息的评估结果;
看客领域更新模块,设置为基于至少一个领域分类中打标信息的评估结果更新所述看客领域库中看客端的看客领域。
在上述实施例的基础上,该装置还包括:
标记设置模块,设置为基于至少一个看客端的打标信息对所述至少一个直播数据设置标记,并将所述至少一个直播数据对应的标记更新至打标库;
视频推荐模块,设置为接收客户端的发送的召回请求,基于所述召回请求中的标记和/或用户画像,以及所述打标库中所述至少一个直播数据的标记,确定推荐直播数据,将所述推荐直播数据发送至所述客户端。
本申请实施例所提供的直播平台可执行本申请任意实施例所提供的直播处理方法,具备执行直播处理方法相应的功能模块。
在上述实施例的基础上,还提供了一个优先示例,参见图6,图6是本申请实施例提供的一种直播平台的结构示意图。该直播平台中包括:登录模块100、看客领域库200、看客领域查询模块300、直播领域确定模块400、推荐模块500、打标评估模块600。
登录模块100设置为在用户请求时进行登录判断。如果未登录则跳转至系统登录模块进行登录,登录完成后重新请求。用户登录后进行权限判断,与看客有关的权限有两种:无看客权限(页面提示注册看客)、有看客权限。无看客权限时提示用户可以在平台上注册成为看客。注册成功后具有看客权限。
看客领域库200包括存储单元,存储单元中包括数据正结构和数据反结构,其中,数据正结构为redis-set结构,key为领域分类id,value为看客id集合,数据反结构为redis-zset结构,key为看客id,value为分类id集合,score为看 客在该分类的权威评分值。看客领域库200中建立正排索引,其查询逻辑为:通过分类id可以查询所有看客的集合,以及建立倒排索引,其查询逻辑为:通过看客id可以查询其擅长的分类集合,并可以看到每个分类的权威评分值。看客领域库200可执行查询操作、新增操作和更新操作,其中查询操作通过redis-zset.zrevrange方法查询排名前N的分类,需要1个参数:看客id;新增操作通过redis.zset.zadd方法向集合添加元素,需要三个参数:看客id,分类id,分类评分;更新操作通过redis.zset.zadd方法进行更新。
看客领域查询模块300设置为匹配看客最擅长的领域,将看客最擅长的领域与直播间匹配后进行推荐。看客通过权限验证后获取看客标识,即看客id,通过看客id,查询看客排名前3的分类id。此时有三种情况:查询结果为0,查询结果为1,查询结果为2。其中,查询结果为0时,说明看客未登记过擅长领域,此时执行初始化逻辑:向看客端展示领域分类信息,提示看客选择最擅长的3个,接收看客端看客根据用户的判断选择的3个擅长的领域(分类id),将记录写入看客领域库。其中key为看客id,value为分类id,score均写为0。查询结果为1时,说明看客只进行过初始录入,将结果标记为low。查询结果为2时,说明看客被智能识别权限领域,将结果标记为high。
直播领域确定模块400设置为获取正在直播的所有直播间的主题风格,将直播推荐给看客。直播开始时,基于直播平台获取直播商品,从而获取商品的类目。使用redis-set结构进行数据存储,key为分类id,value为直播id,直播结束时,回调服务端,将直播id从set中移除。其中一场直播可能有多个品,但是主分类id一般相同。
推荐模块500设置为将直播间和看客端进行映射匹配,为看客端推荐合适的直播间,示例性的,参见图7,图7是本申请实施例提供的直播间推荐的示意图。看客领域查询模块300查询结果为2,即结果标记为high时,看客端经过智能识别擅长类目是a1,a2,a3。选取4个类目,其中3个为看客的擅长类目,1个为不包括前三个类目的随机类目,为了在应用看客的最优能力和发掘潜在能 力之间达到平衡。随机挑选上述4个类目之一,用该领域分类标识即分类id查询生成的直播间id推送给看客端,看客侧看到的表现为下滑画面时出现一个新的直播。
看客领域查询模块300查询结果为1,即结果标记为low时,看客端未经过智能识别自选类目为a1,a2。选取4个类目,其中2个为看客选择的擅长类目记为集合A,2个位不包括前两个类目的随机类目记为集合B。每次查询时随机从集合A和集合B各挑选一个类目id,使用上述两个类目id,分别查询一个直播间id,将这个直播间id推送给看客端,看客侧看到的表现为下滑画面时出现选项,请看客选择观看哪个直播间,看客点击后进入对应直播间,以更准确的为看客做出初始识别。
打标评估模块600设置为对看客权威领域进行自动更新,以更好的识别看客,从而推荐更匹配的直播间给看客。看客端针对直播间进行观看、诊断、保存诊断结果,看客诊断成功时,获取看客诊断成功的分类id,对看客领域库200进行更新,其中key为看客id,value为分类id,score为原始值+1,如果无记录,则score默认值为0。看客每次的诊断结果,将更新其对应的领域库,便于下次推荐更合适的直播间,从而实现推荐能力的自动升级。
实施例五
图8是本申请实施例五提供的一种直播处理装置的结构示意图,该直播处理装置可以是配置在看客端。该装置包括:
第一数据请求模块510,设置为向直播平台发送召回请求,并接收与当前设备的看客领域相匹配的推荐打标直播的直播标识;
第二数据请求模块520,设置为基于选择的推荐打标直播的直播标识生成直播数据请求,将所述直播数据请求发送至所述直播平台,并接收所述直播平台反馈的直播数据以及待打标信息;
打标模块530,设置为显示所述直播数据以及所述待打标信息,并获取所述 待打标信息对应的打标信息;
打标信息反馈模块540,设置为将所述打标信息以及所述直播数据对应的直播标识发送至所述直播平台,以使所述直播平台存储所述直播数据的打标信息,其中,所述打标信息为所述直播平台中对应所述直播数据的推荐参考信息。
在上述实施例的基础上,该装置还包括:
领域分类信息展示模块,设置为接收所述直播平台反馈的领域分类信息,并将所述领域分类信息在显示界面进行显示;
领域分类信息反馈模块,设置为检测用户输入的对领域分类信息的选择信息,将选择的领域分类信息发送至所述直播平台,其中,所述选择的领域分类信息用于确定看客领域。
在上述实施例的基础上,打标信息反馈模块540设置为:
显示所述直播数据,并进行直播显示计时;
若所述直播显示计时满足预设显示时长,则显示所述待打标信息。
本申请实施例所提供的直播处理装置可执行本申请任意实施例所提供的直播处理方法,具备执行直播处理方法相应的功能模块。
实施例六
图9是本申请实施例六提供的一种直播系统的结构示意图,该直播系统包括:
直播端610,设置为获取至少一个直播数据,并将至少一个直播数据发送至直播平台;
看客端620,设置为向直播平台发送召回请求;
直播平台630,设置为基于看客端的看客标识确定看客端的看客领域,获取所述至少一个直播数据的当前直播领域,基于看客领域以及所述至少一个直播数据的当前直播领域确定看客端的推荐打标直播,将推荐打标直播发送至看客端;
看客端620接收推荐打标直播,展示推荐打标直播,并采集对推荐打标直播的打标信息,将打标信息发送至直播平台;
直播平台630还设置为接收看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,打标信息为对应直播数据的推荐参考信息。
在上述实施例的基础上,直播系统还包括客户端,客户端与直播平台通信连接,客户端设置为向所述直播平台发送召回请求;
直播平台,还设置为接收至少一个看客端反馈的打标信息,基于所述至少一个看客端的打标信息对所述至少一个直播数据设置标记,并将至少一个直播数据对应的标记更新至打标库;
以及,接收客户端的发送的召回请求,基于所述召回请求中的标记和/或用户画像,以及所述打标库中所述至少一个直播数据的标记,确定推荐直播数据,将所述推荐直播数据发送至所述客户端。
本实施例的技术方案,通过在当前的至少一个直播数据中,基于看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播,将所述推荐打标直播的直播标识发送至所述看客端,以使看客端对推荐打标直播进行打标。由于推荐打标直播是根据看客领域针对性确定的,在看客端的擅长领域范围内,有利于看客端针对所擅长领域内的直播数据进行打标,提高打标信息的正确率和打标效率。同时将至少一个直播数据的打标信息进行存储,便于在客户端请求直播数据时,基于打标信息向客户端进行精准推荐,提高直播数据的推荐精准度。
实施例七
图10为本申请实施例七提供的一种电子设备的结构示意图。图10示出了适于用来实现本申请实施方式的电子设备12的框图。图10显示的电子设备12仅仅是一个示例,不应对本申请实施例的功能和使用范围带来任何限制。设备12典型的是承担图像分类功能的电子设备。
如图10所示,电子设备12以通用计算设备的形式表现。电子设备12的组件可以包括但不限于:至少一个处理器16,存储装置28,连接不同系统组件(包括存储装置28和处理器16)的总线18。
总线18表示几类总线结构中的至少一种,包括存储器总线或者存储器控制器,外围总线,图形加速端口,处理器或者使用多种总线结构中的任意总线结构的局域总线。举例来说,这些体系结构包括但不限于工业标准体系结构(Industry Standard Architecture,ISA)总线,微通道体系结构(Micro Channel Architecture,MCA)总线,增强型ISA总线、视频电子标准协会(Video Electronics Standards Association,VESA)局域总线以及外围组件互连(Peripheral Component Interconnect,PCI)总线。
电子设备12典型地包括多种计算机系统可读介质。这些介质可以是任何能够被电子设备12访问的可用介质,包括易失性和非易失性介质,可移动的和不可移动的介质。
存储装置28可以包括易失性存储器形式的计算机系统可读介质,例如随机存取存储器(Random Access Memory,RAM)30和/或高速缓存存储器32。电子设备12可以包括其它可移动/不可移动的、易失性/非易失性计算机系统存储介质。仅作为举例,存储系统34可以设置为读写不可移动的、非易失性磁介质(图10未显示,通常称为“硬盘驱动器”)。尽管图10中未示出,可以提供用于对可移动非易失性磁盘(例如“软盘”)读写的磁盘驱动器,以及对可移动非易失性光盘(例如只读光盘(Compact Disc-Read Only Memory,CD-ROM)、数字视盘(Digital Video Disc-Read Only Memory,DVD-ROM)或者其它光介质)读写的光盘驱动器。在这些情况下,每个驱动器可以通过至少一个数据介质接口与总线18相连。存储装置28可以包括至少一个程序产品,该程序产品具有一组(例如至少一个)程序模块,这些程序模块被配置以执行本申请各实施例的功能。
具有一组(至少一个)程序模块26的程序36,可以存储在例如存储装置 28中,这样的程序模块26包括但不限于操作系统、至少一个应用程序、其它程序模块以及程序数据,这些示例中的每一个或某种组合中可能包括网关环境的实现。程序模块26通常执行本申请所描述的实施例中的功能和/或方法。
电子设备12也可以与至少一个外部设备14(例如键盘、指向设备、摄像头、显示器24等)通信,还可与至少一个使得用户能与该电子设备12交互的设备通信,和/或与使得该电子设备12能与至少一个其它计算设备进行通信的任何设备(例如网卡,调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口22进行。并且,电子设备12还可以通过网关适配器20与至少一个网关(例如局域网(Local Area Network,LAN),广域网Wide Area Network,WAN)和/或公共网关,例如因特网)通信。如图所示,网关适配器20通过总线18与电子设备12的其它模块通信。应当明白,尽管图中未示出,可以结合电子设备12使用其它硬件和/或软件模块,包括但不限于:微代码、设备驱动器、冗余处理单元、外部磁盘驱动阵列、磁盘阵列(Redundant Arrays of Independent Disks,RAID)系统、磁带驱动器以及数据备份存储系统等。
处理器16通过运行存储在存储装置28中的程序,从而执行各种功能应用以及数据处理,例如实现本申请上述实施例所提供的直播处理方法。
实施例八
本申请实施例八提供了一种计算机可读存储介质,其上存储有计算机程序,该程序被处理器执行时实现如本申请实施例所提供的直播处理方法。
当然,本申请实施例所提供的一种计算机可读存储介质,其上存储的计算机程序不限于如上所述的方法操作,还可以执行本申请任意实施例所提供的直播处理方法。
本申请实施例的计算机存储介质,可以采用至少一个计算机可读的介质的任意组合。计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质。计算机可读存储介质例如可以是——但不限于——电、磁、光、电磁、红 外线、或半导体的系统、装置或器件,或者任意以上的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括:具有至少一个导线的电连接、便携式计算机磁盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器((Erasable Programmable Read-Only Memory,EPROM)或闪存)、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本文件中,计算机可读存储介质可以是任何包含或存储程序的有形介质,该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
计算机可读的信号介质可以包括在基带中或者作为载波一部分传播的数据信号,其中承载了计算机可读的源代码。这种传播的数据信号可以采用多种形式,包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读的信号介质还可以是计算机可读存储介质以外的任何计算机可读介质,该计算机可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。
计算机可读介质上包含的源代码可以用任何适当的介质传输,包括——但不限于无线、电线、光缆、射频(Radio Frequency,RF)等等,或者上述的任意合适的组合。
可以以一种或多种程序设计语言或其组合来编写用于执行本申请操作的计算机源代码,所述程序设计语言包括面向对象的程序设计语言—诸如Java、Smalltalk、C++,还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。源代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中,远程计算机可以通过任意种类的网关——包括局域网(LAN)或广域网(WAN)—连接到用户计算机,或者,可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。

Claims (17)

  1. 一种直播处理方法,包括:
    获取看客端的召回请求,其中,所述看客端设置为对推荐直播进行打标;
    获取至少一个直播数据的当前直播领域,基于所述看客端的看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播;
    将所述推荐打标直播的直播标识发送至所述看客端,并接收所述看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,所述打标信息为对应直播数据的推荐参考信息。
  2. 根据权利要求1所述的方法,在基于所述看客端的看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播之前,还包括:
    将所述看客端的看客标识在预设的看客领域库中进行匹配,确定所述看客标识对应的至少一个看客领域;或者,
    响应于所述看客领域库中不包括所述看客标识,向看客端发送领域分类信息,接收所述看客端反馈的看客领域,并更新所述看客领域库。
  3. 根据权利要求1所述的方法,其中,所述获取至少一个直播数据的当前直播领域,包括:
    对于任一直播数据,基于所述任一直播数据中的核心直播对象确定所述任一直播数据的当前直播领域,或者,基于所述任一直播数据中的当前直播对象确定所述任一直播数据的当前直播领域。
  4. 根据权利要求1所述的方法,其中,所述看客端的推荐打标直播包括:所述看客领域对应的第一直播和所述看客领域以外的其他领域分类的第二直播,所述第一直播和所述第二直播的数量满足预设比例。
  5. 根据权利要求4所述的方法,其中,所述第一直播和所述第二直播的数量满足的预设比例基于所述看客端的看客领域识别方式确定,其中,所述看客领域识别方式包括看客端输入和打标评估识别。
  6. 根据权利要求1所述的方法,其中,所述基于所述看客领域以及至少一个所述直播数据的当前直播领域确定所述看客端的推荐打标直播,包括:
    根据所述至少一个直播数据的直播状态,确定至少一个可推荐直播数据;
    基于所述看客领域以及所述至少一个可推荐直播数据确定所述看客端的推荐打标直播。
  7. 根据权利要求6所述的方法,其中,所述直播状态包括当前直播对象的直播时长,所述推荐打标直播的数量为至少一个;
    所述方法还包括:
    基于至少一个推荐打标直播的直播状态确定所述至少一个推荐打标直播的推荐优先级。
  8. 根据权利要求2所述的方法,在接收所述看客端反馈的打标信息之后,所述方法还包括:
    对所述打标信息进行评估,确定所述打标信息的评估结果;
    基于至少一个领域分类中打标信息的评估结果更新所述看客领域库中看客端的看客领域。
  9. 根据权利要求1所述的方法,其中,所述看客端的数量为至少一个,所述方法还包括:
    基于至少一个看客端的打标信息对所述所述至少一个直播数据设置标记,并将所述至少一个直播数据对应的标记更新至打标库;
    接收客户端的发送的召回请求,基于所述召回请求中的标记和用户画像中的至少之一,以及所述打标库中所述至少一个直播数据的标记,确定推荐直播数据,将所述推荐直播数据发送至所述客户端。
  10. 一种直播处理方法,包括:
    向直播平台发送召回请求,并接收与当前设备的看客领域相匹配的推荐打标直播的直播标识;
    基于选择的推荐打标直播的直播标识生成直播数据请求,将所述直播数据请求发送至所述直播平台,并接收所述直播平台反馈的直播数据以及待打标信息;
    显示所述直播数据以及所述待打标信息,并获取所述待打标信息对应的打标信息;
    将所述打标信息以及所述直播数据对应的直播标识发送至所述直播平台,以使所述直播平台存储所述直播数据的打标信息,其中,所述打标信息为所述直播平台中对应所述直播数据的推荐参考信息。
  11. 根据权利要求10所述的方法,在向直播平台发送召回请求之后,所述方法还包括:
    接收所述直播平台反馈的领域分类信息,并将所述领域分类信息在显示界面进行显示;
    检测用户输入的对领域分类信息的选择信息,将选择的领域分类信息发送至所述直播平台,其中,所述选择的领域分类信息用于确定看客领域。
  12. 根据权利要求10所述的方法,其中,所述显示所述直播数据以及所述待打标信息,包括:
    显示所述直播数据,并进行直播显示计时;
    响应于所述直播显示计时满足预设显示时长,显示所述待打标信息。
  13. 一种直播平台,包括:
    召回请求获取模块,设置为获取看客端的召回请求,其中,所述看客端设置为对推荐直播进行打标;
    当前直播领域获取模块,设置为获取至少一个直播数据的当前直播领域;
    推荐打标直播确定模块,设置为基于所述看客端的看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播;
    打标信息获取模块,设置为将所述推荐打标直播的直播标识发送至所述看客端,并接收所述看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,所述打标信息为对应直播数据的推荐参考信息。
  14. 一种直播处理装置,包括:
    第一数据请求模块,设置为向直播平台发送召回请求,并接收与当前设备 的看客领域相匹配的推荐打标直播的直播标识;
    第二数据请求模块,设置为基于选择的推荐打标直播的直播标识生成直播数据请求,将所述直播数据请求发送至所述直播平台,并接收所述直播平台反馈的直播数据以及待打标信息;
    打标模块,设置为显示所述直播数据以及所述待打标信息,并获取所述待打标信息对应的打标信息;
    打标信息反馈模块,设置为将所述打标信息以及所述直播数据对应的直播标识发送至所述直播平台,以使所述直播平台存储所述直播数据的打标信息,其中,所述打标信息为所述直播平台中对应所述直播数据的推荐参考信息。
  15. 一种直播系统,包括:
    直播端,设置为获取至少一个直播数据,并将所述至少一个直播数据发送至直播平台;
    看客端,设置为向直播平台发送召回请求;
    所述直播平台,设置为获取所述至少一个直播数据的当前直播领域,基于所述看客端的看客领域以及所述至少一个直播数据的当前直播领域确定所述看客端的推荐打标直播,将所述推荐打标直播的直播标识发送至所述看客端;
    所述看客端接收所述推荐打标直播,展示所述推荐打标直播,并采集对所述推荐打标直播的打标信息,将所述打标信息发送至所述直播平台;
    所述直播平台还设置为接收所述看客端反馈的打标信息,存储所述至少一个直播数据的打标信息,其中,所述打标信息为对应直播数据的推荐参考信息。
  16. 一种电子设备,包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行所述计算机程序时实现如权利要求1-9中任一所述的直播处理方法,或者,权利要求10-12中任一所述的直播处理方法。
  17. 一种计算机可读存储介质,所述计算机可读存储介质上存储有计算机程序,所述计算机程序被处理器执行时实现如权利要求1-9中任一所述的直播处理方法,或者,权利要求10-12中任一所述的直播处理方法。
PCT/CN2022/095233 2021-05-28 2022-05-26 直播处理方法、直播平台、装置、系统、介质及设备 WO2022247906A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110594732.8 2021-05-28
CN202110594732.8A CN113315989B (zh) 2021-05-28 2021-05-28 直播处理方法、直播平台、装置、系统、介质及设备

Publications (1)

Publication Number Publication Date
WO2022247906A1 true WO2022247906A1 (zh) 2022-12-01

Family

ID=77376331

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/095233 WO2022247906A1 (zh) 2021-05-28 2022-05-26 直播处理方法、直播平台、装置、系统、介质及设备

Country Status (2)

Country Link
CN (1) CN113315989B (zh)
WO (1) WO2022247906A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113315989B (zh) * 2021-05-28 2022-12-27 北京沃东天骏信息技术有限公司 直播处理方法、直播平台、装置、系统、介质及设备
CN113992934B (zh) * 2021-11-01 2024-01-30 北京字跳网络技术有限公司 多媒体信息处理方法、装置、电子设备和存储介质
CN114422817B (zh) * 2022-01-17 2023-09-26 北京达佳互联信息技术有限公司 直播间数据处理方法、装置、电子设备、存储介质

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106792081A (zh) * 2016-12-07 2017-05-31 腾讯科技(深圳)有限公司 直播视频的推送方法及装置
US20180150465A1 (en) * 2016-11-30 2018-05-31 Facebook, Inc. Systems and methods for providing content
CN108271076A (zh) * 2017-01-03 2018-07-10 武汉斗鱼网络科技有限公司 一种推荐直播间的方法及装置
CN109063133A (zh) * 2018-08-02 2018-12-21 武汉斗鱼网络科技有限公司 一种直播间标签的添加方法、系统、设备及介质
CN109104639A (zh) * 2017-06-20 2018-12-28 阿里巴巴集团控股有限公司 直播系统、确定直播视频主题的方法、装置以及电子设备
CN109547863A (zh) * 2018-10-22 2019-03-29 武汉斗鱼网络科技有限公司 一种标签的标记方法、装置、服务器和存储介质
CN110602514A (zh) * 2019-09-12 2019-12-20 腾讯科技(深圳)有限公司 一种直播频道的推荐方法、装置、电子设备及存储介质
CN112565828A (zh) * 2020-12-17 2021-03-26 大兴安岭林海明珠网络技术服务有限公司 一种直播视频推送方法
CN113099260A (zh) * 2021-04-21 2021-07-09 北京沃东天骏信息技术有限公司 直播处理方法、直播平台、系统、介质及电子设备
CN113315989A (zh) * 2021-05-28 2021-08-27 北京沃东天骏信息技术有限公司 直播处理方法、直播平台、装置、系统、介质及设备

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9762943B2 (en) * 2015-11-16 2017-09-12 Telefonaktiebolaget Lm Ericsson Techniques for generating and providing personalized dynamic live content feeds
CN111079012A (zh) * 2019-12-13 2020-04-28 广州酷狗计算机科技有限公司 直播间推荐方法、装置、存储介质及终端
CN111866528A (zh) * 2020-04-30 2020-10-30 火币(广州)区块链科技有限公司 一种直播节目推送方法和可读存储介质
CN112770126A (zh) * 2020-12-29 2021-05-07 北京达佳互联信息技术有限公司 直播间的推送方法、装置、服务器及存储介质

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180150465A1 (en) * 2016-11-30 2018-05-31 Facebook, Inc. Systems and methods for providing content
CN106792081A (zh) * 2016-12-07 2017-05-31 腾讯科技(深圳)有限公司 直播视频的推送方法及装置
CN108271076A (zh) * 2017-01-03 2018-07-10 武汉斗鱼网络科技有限公司 一种推荐直播间的方法及装置
CN109104639A (zh) * 2017-06-20 2018-12-28 阿里巴巴集团控股有限公司 直播系统、确定直播视频主题的方法、装置以及电子设备
CN109063133A (zh) * 2018-08-02 2018-12-21 武汉斗鱼网络科技有限公司 一种直播间标签的添加方法、系统、设备及介质
CN109547863A (zh) * 2018-10-22 2019-03-29 武汉斗鱼网络科技有限公司 一种标签的标记方法、装置、服务器和存储介质
CN110602514A (zh) * 2019-09-12 2019-12-20 腾讯科技(深圳)有限公司 一种直播频道的推荐方法、装置、电子设备及存储介质
CN112565828A (zh) * 2020-12-17 2021-03-26 大兴安岭林海明珠网络技术服务有限公司 一种直播视频推送方法
CN113099260A (zh) * 2021-04-21 2021-07-09 北京沃东天骏信息技术有限公司 直播处理方法、直播平台、系统、介质及电子设备
CN113315989A (zh) * 2021-05-28 2021-08-27 北京沃东天骏信息技术有限公司 直播处理方法、直播平台、装置、系统、介质及设备

Also Published As

Publication number Publication date
CN113315989A (zh) 2021-08-27
CN113315989B (zh) 2022-12-27

Similar Documents

Publication Publication Date Title
WO2022247906A1 (zh) 直播处理方法、直播平台、装置、系统、介质及设备
US10945043B2 (en) Methods and apparatus to create a panel of media device users
CN108702524B (zh) 利用多匹配检测和基于位置的消歧的媒体频道识别
US10235025B2 (en) Various systems and methods for expressing an opinion
US10698906B2 (en) Electronic data generation methods
US9516366B2 (en) Determining user engagement with media content via mobile device usage
WO2021143350A1 (zh) 数据处理方法及直播系统
ES2358219T3 (es) Sistema de transmisión de informaciones, sistema de transmisión pública de informaciones multimedia y su método de transmisión.
CN108848393B (zh) 入口展示方法、装置、设备及存储介质
CN108702543A (zh) 基于音频指纹的视频多匹配检测和对媒体频道识别消歧
CN108702525B (zh) 利用多匹配检测和基于单匹配的歧义消除的媒体频道识别
WO2017219527A1 (zh) 用于智能终端的数据统计分析方法及系统
WO2022247220A9 (zh) 界面处理方法及装置
WO2022193071A1 (zh) 一种基于直播的送礼方法及装置、设备以及存储介质
CN111417021B (zh) 外挂识别方法、装置、计算机设备和可读存储介质
JP5243136B2 (ja) 情報処理装置、情報処理システム、情報処理方法、およびプログラム
CN106127516A (zh) 一种统计广告投放数据的方法及装置
JP5243137B2 (ja) 情報処理装置、情報処理システム、情報処理方法、およびプログラム
CN113099260B (zh) 直播处理方法、直播平台、系统、介质及电子设备
JP6026454B2 (ja) コミュニティ提供システム、コミュニティ提供方法及びコミュニティ提供プログラム
CN109348249A (zh) 确定数字专辑的用户获取数量的方法、装置及存储介质
CN104427361A (zh) 电视服务系统与提供影音服务的方法
RU2703968C1 (ru) Способ и система для интерактивной таргетированной видеорекламы на основе истории пользовательских сессий
AU2013204449B2 (en) Electronic data generation methods
CN116801013A (zh) 一种显示设备、服务器及媒资数据的显示方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22810630

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE