Data processing method, apparatus, device, and medium

Abstract

The application discloses a data processing method and device, equipment and medium, and relates to the field of data processing. The method comprises the following steps: obtaining the scroll data of n videos, wherein the n videos belong to the same series; determining at least one influence parameter of an i-th video in the n videos based on the scroll data, wherein the influence parameter is used for indicating the factor importance parameter corresponding to the quantitative index of the i-th video; determining the playing influence parameter associated with the i-th video based on the at least one influence parameter, wherein the playing influence parameter is used for indicating the factor importance parameter corresponding to the non-quantitative index of the i-th video; and predicting the abnormal influence degree of the non-quantitative index on future video playing according to n playing influence parameters corresponding to the n videos respectively, and determining the playing strategy of the future video playing.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to a data processing method, apparatus, device and medium. Background Technology

[0002] With the development of the esports industry, esports event live streaming has gained widespread attention, and organizers of esports events need to rely on live streaming platforms to broadcast the events. Typically, esports viewership data is controlled by the live streaming platforms, and event organizers cannot obtain accurate viewership data, thus preventing them from adjusting future event broadcasts. Summary of the Invention

[0003] This application provides a data processing method, apparatus, device, and medium, offering a way to process bullet screen data. This enables event organizers to determine future livestream playback strategies based on bullet screen data from livestreams, thereby allowing for adjustments to future livestreams. The technical solution is as follows:

[0004] According to one aspect of this application, a data processing method is provided, the method comprising:

[0005] Get the bullet comment data for n videos, where the n videos belong to the same series;

[0006] Based on the bullet screen data, determine at least one influence parameter for the i-th video among n videos. The influence parameter is used to indicate the factor importance parameter corresponding to the quantitative indicator for playing the i-th video.

[0007] Based on at least one influence parameter, determine the playback influence parameter associated with the i-th video. The playback influence parameter is used to indicate the factor importance parameter corresponding to the non-quantitative indicator for playing the i-th video.

[0008] Based on the n playback impact parameters corresponding to each of the n videos, predict the degree of abnormal impact of non-quantitative indicators on future video playback, and determine the playback strategy for future video playback.

[0009] Where n and i are positive integers greater than zero, and n is not less than i.

[0010] According to one aspect of this application, a data processing apparatus is provided, the apparatus comprising:

[0011] The acquisition module is used to acquire the bullet comment data of n videos, where the n videos belong to the same series;

[0012] The processing module is used to determine at least one influence parameter of the i-th video among n videos based on the bullet screen data. The influence parameter is used to indicate the factor importance parameter corresponding to the quantitative indicator for playing the i-th video.

[0013] The processing module is also used to determine the playback influence parameter associated with the i-th video based on at least one influence parameter. The playback influence parameter is used to indicate the factor importance parameter corresponding to the non-quantitative indicator for playing the i-th video.

[0014] The processing module is also used to predict the degree of abnormal impact of non-quantitative indicators on future video playback based on the n playback impact parameters corresponding to the n videos respectively, and to determine the playback strategy for future video playback.

[0015] Where n and i are positive integers greater than zero, and n is not less than i.

[0016] According to one aspect of this application, a computer device is provided, the computer device including a memory and a processor; the memory stores at least one piece of program code, which is loaded and executed by the processor to implement the data processing method described above.

[0017] According to one aspect of this application, a computer-readable storage medium is provided, in which a computer program is stored, the computer program being executed by a processor to implement the data processing method described above.

[0018] According to one aspect of this application, a chip is provided, the chip including programmable logic circuitry and / or program instructions, for implementing the data processing method described above when an electronic device on which the chip is mounted is running.

[0019] According to one aspect of this application, a computer program product is provided, comprising computer instructions stored in a computer-readable storage medium, wherein a processor reads from and executes the computer instructions to implement the data processing method described above.

[0020] The beneficial effects of the technical solutions provided in this application include at least the following:

[0021] A novel data processing method is provided. By processing the bullet comment data of n videos in the same series, the playback impact parameters associated with each video can be determined. Based on the playback impact parameters, the influence of non-quantitative indicators on future video playback can be predicted to determine the playback strategy for future video playback. This allows the organizers of the video playback to adjust the future video playback based on the bullet comment data. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of a computer system provided in an exemplary embodiment of this application;

[0024] Figure 2 This is a flowchart of a data processing method provided in an exemplary embodiment of this application;

[0025] Figure 3 This is a flowchart illustrating the acquisition of bullet screen data provided in an exemplary embodiment of this application;

[0026] Figure 4 This is a flowchart of a data processing method provided in an exemplary embodiment of this application;

[0027] Figure 5 This is a flowchart of a data processing method provided in an exemplary embodiment of this application;

[0028] Figure 6 This is a flowchart of a data processing method provided in an exemplary embodiment of this application;

[0029] Figure 7 This is a schematic diagram illustrating the relationship between the live stream engagement parameter and the average number of bullet comments provided in an exemplary embodiment of this application;

[0030] Figure 8 This is a flowchart of a data processing method provided in an exemplary embodiment of this application;

[0031] Figure 9 This is a schematic diagram of a data processing apparatus provided in an exemplary embodiment of this application;

[0032] Figure 10 This is a schematic diagram of the structure of a computer device provided in an exemplary embodiment of this application. Detailed Implementation

[0033] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0034] It should be understood that "several" in this article refers to one or more, and "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0035] Video streaming platforms typically feature video series, such as live quarterly tournament broadcasts, drama series, and documentary series. The playback of n videos belonging to the same series is influenced by multiple factors. This application provides a data processing method that can determine the playback strategy for future videos by processing the bullet screen data of n videos.

[0036] Taking live esports broadcasts as an example, esports competitions can be presented via live streaming. Organizers of esports events typically rely on live streaming platforms to broadcast the events. The platform provides viewable live videos to user accounts, who can then log in and watch the stream. However, esports viewership data is usually controlled by the live streaming platform, preventing the event organizers from accessing accurate viewership data and thus hindering their ability to adjust future broadcasts.

[0037] Based on this, the bullet screen data of multiple live broadcasts of historical matches can be analyzed and processed to obtain multiple influencing parameters for each live broadcast, thereby determining the non-quantitative indicators of the match and predicting the impact of the non-quantitative indicators on future live broadcasts, so as to determine the corresponding playback strategy.

[0038] In illustrative terms, bullet screen data is used to indicate the data information associated with bullet screens displayed on the video playback platform's display interface, including but not limited to quantity information, text information, etc.

[0039] Taking TV series as an example, the production company can broadcast its completed episodes on a certain platform. Similar to the previous example, the production company also cannot obtain the actual playback data, thus preventing them from adjusting the playback of subsequent episodes.

[0040] Based on this, the bullet screen data of historical dramas that have already been played can be analyzed and processed to obtain the impact parameters of each episode, thereby determining non-quantitative indicators, predicting the impact of non-quantitative indicators on future video playback, and determining the corresponding playback strategy.

[0041] It should be understood that the various embodiments given in this application all take live sports events as an example, but live sports events do not limit this application, and the data processing methods provided in the embodiments of this application can be applied to other series of videos.

[0042] First, a brief introduction to the terms used in the embodiments of this application will be given.

[0043] Event: A competition consisting of two different teams competing against each other.

[0044] In some embodiments, a schedule may include at least one event, or a schedule may include multiple events. For example, within the same schedule (such as a quarterly season or regular season), there are at least two teams, and at least one event is composed of two of these two teams. For instance, in a certain schedule, there are 8 teams. These 8 teams are drawn to form 4 events; the winners of the 4 events are drawn to form 2 more events, and the losers are eliminated; the winners of the subsequent 2 events form a championship / runner-up match, and the losers form a third-place match. Based on this, the schedule includes 8 events, and the teams competing in each event are different, with each team participating in a different number of events.

[0045] Bullet screen data: Data information associated with bullet screens displayed on the display interface of a video playback platform when a video is played.

[0046] The bullet screen data can be result-type data such as average bullet screen count, average bullet screen count per minute, and total bullet screen count, or it can be the text data carried in the bullet screen. For illustration, the average bullet screen count indicates the number of bullet screens generated by users watching the video per unit of time; the average bullet screen count per minute indicates the average number of bullet screens generated by users watching the video per minute, and can be considered a type of average bullet screen count; the total bullet screen count indicates the total number of bullet screens generated by users watching the video during the entire video playback.

[0047] Figure 1 The diagram illustrates a computer system 100 provided in an exemplary embodiment of this application. The computer system 100 includes a data processing device 110 and a data capture device 120. The data processing device 110 is used to process bullet screen data, and the data capture device 120 is used to acquire, process, and store bullet screen data displayed on the live streaming platform's interface.

[0048] In some embodiments, bullet comments can be obtained through bullet comment capture technology, such as obtaining bullet comments from the live streaming platform in real time through Transmission Control Protocol (TCP) technology, or obtaining bullet comments from the live streaming platform in real time through TCP-based full-duplex communication protocol (Websocket) technology.

[0049] After acquiring the bullet screen data, the data capture device 120 can directly send the bullet screen data to the data processing device 110, or it can perform simple processing on the bullet screen data before sending it to the data processing device 110. This application does not limit this.

[0050] The subsequent processing of the bullet screen data by the data processing device 110 will be described below, and will be omitted here.

[0051] Optionally, the data processing device 110 and the data capturing device 120 may be the same computer device, or they may be different computer devices. Furthermore, when the data processing device 110 and the data capturing device 120 are different devices, they may be of the same type, such as both being servers; or they may be of different types, such as the data processing device 110 being a server and the data capturing device 120 being a terminal.

[0052] The aforementioned servers can be independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. The aforementioned terminals can be smartphones, tablets, laptops, desktop computers, smart TVs, in-vehicle terminals, wearable devices, and smart speakers, but are not limited to these. Terminals and servers can be directly or indirectly connected via wired or wireless communication, and this application does not impose any restrictions on this connection.

[0053] It should be noted that this application can display a prompt interface, pop-up window, or output voice prompt information before and during the acquisition of bullet screen data. The prompt interface, pop-up window, or voice prompt information is used to inform the user that their relevant data is being collected. This ensures that this application only begins to execute the relevant steps for acquiring user data after receiving confirmation from the user regarding the prompt interface or pop-up window. Otherwise (i.e., if no confirmation from the user is received regarding the prompt interface or pop-up window), the relevant steps for acquiring user data are terminated, and the user's relevant data is not acquired.

[0054] It should be understood that all user data collected in this application is collected with the user's consent and authorization, and the collection, use and processing of relevant user data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0055] Figure 2 A flowchart of a data processing method provided in an exemplary embodiment of this application is shown. The method is composed of... Figure 1 The data processing device 110 in the middle executes the method, which includes the following steps:

[0056] Step 102: Obtain the bullet comment data for n videos.

[0057] This is illustrative, showing n videos belonging to the same series. Referring to the preceding content, n videos belonging to the same series are related, such as live quarterly tournament broadcasts, drama series, and documentary series.

[0058] Taking live esports broadcasts as an example, esports competitions can be presented via live streaming. Organizers of esports events typically rely on live streaming platforms to broadcast the events. The platform provides viewable live videos to user accounts, who can log in and watch the currently streaming video. For instance, n videos belong to the live streams of n matches within the same tournament schedule. A tournament schedule includes at least one match, each match has one live stream, and each match involves two teams competing.

[0059] Taking a TV series as an example, the production company can broadcast its completed episodes on a certain platform. Similar to the previous example, the production company also cannot obtain actual playback data, thus preventing them from adjusting the playback of subsequent episodes. For instance, n videos belong to n episodes of the same TV series.

[0060] To illustrate, when a video is played on a certain video playback platform, the bullet screen data is used to indicate the data information associated with the bullet screen displayed on the display interface of the video playback platform during each video playback overshoot.

[0061] The bullet screen data can be result-type data such as average bullet screen count, average bullet screen count per minute, and total bullet screen count, or it can be the text data carried in the bullet screen. For illustration, the average bullet screen count indicates the number of bullet screens generated by users watching the video per unit of time; the average bullet screen count per minute indicates the average number of bullet screens generated by users watching the video per minute, and can be considered a type of average bullet screen count; the total bullet screen count indicates the total number of bullet screens generated by users watching the video during the entire video playback.

[0062] Figure 3A flowchart illustrating the acquisition of bullet screen data provided in an exemplary embodiment of this application is shown.

[0063] The bullet screen data can be obtained through bullet screen capture technology, such as real-time acquisition using TCP / Websocket technology. After acquiring the bullet screen data, the device can receive the data in real time and perform two different processing steps on it.

[0064] On the one hand, the acquired raw data (i.e., the raw bullet screen data obtained through bullet screen capture technology) can be stored asynchronously. For example, a cloud server storage solution can be used to asynchronously back up the raw data, ensuring data security and reliability. On the other hand, the bullet screen data can be calculated in real time to obtain result-type data, including but not limited to the aforementioned average number of bullet screens, average number of bullet screens per minute, and total number of bullet screens, to ensure the real-time nature and validity of the data.

[0065] refer to Figure 3 It should be understood that, based on asynchronously stored raw data, the calculation of result-type data can also be achieved by replaying the stored raw data.

[0066] The following examples all use live sports broadcasts:

[0067] In this context, a tournament refers to a competition event consisting of two different teams. It can be understood as a schedule (such as a quarterly tournament or regular season) containing at least two teams, with at least one tournament consisting of two of these teams. Furthermore, when broadcasting this schedule live, each tournament can be streamed live, generating different chat data for each live stream.

[0068] For example, when live streaming n matches corresponding to n matches of the same schedule on a certain live streaming platform, the bullet screen data refers to the data information associated with the bullet screens displayed on the display interface of each match during the live stream. In some embodiments, the bullet screen data includes average bullet screen count, average bullet screen count per minute, and total bullet screen count. The average bullet screen count indicates the number of bullet screens generated by users watching the live stream per unit of time; the average bullet screen count per minute indicates the average number of bullet screens generated by users watching the live stream per minute, and can be considered a type of average bullet screen count; the total bullet screen count indicates the total number of bullet screens generated by users watching the live stream in this particular match.

[0069] It should be noted that the bullet screen data involved in this application can be raw data or calculated result data. If raw data is used, the data processing method provided in this application embodiment requires additional processing of the raw data. Furthermore, subsequent data processing will cover all bullet screen data corresponding to n videos belonging to the same series. Taking a live sports event as an example, it can be understood that the data processing method provided in this application embodiment is based on all bullet screen data belonging to the same event, using that event as the basis for data processing.

[0070] Step 104: Based on the bullet screen data, determine at least one influencing parameter for the i-th video among the n videos.

[0071] Indicatively, the influence parameter is used to indicate the factor importance parameter corresponding to the quantification index of the i-th video, where n and i are positive integers greater than zero, and n is not less than i.

[0072] Quantifiable metrics refer to quantifiable indicators related to the i-th video. Taking live sports broadcasts as an example, quantifiable metrics include, but are not limited to, metrics associated with the current match, metrics associated with the match's schedule, and metrics involved in broadcasting the match. For example, quantifiable metrics include, but are not limited to, at least one of the following: participating teams, commentators, broadcast duration, schedule attributes, and the time period of the match. It should be understood that all of the above metrics affect the bullet comment data. For example, if a participating team is highly popular, it will lead to an increase in the number of bullet comments sent by users, and most of these comments will be relatively friendly. Conversely, if the commentators have weak communication skills, fewer people will watch the live broadcast, resulting in a decrease in the number of bullet comments.

[0073] Based on this, after obtaining the bullet screen data, and through quantification of the aforementioned indicators, we can obtain the factor importance parameters corresponding to different quantified indicators, that is, different influence parameters. The influence parameters also vary depending on the video.

[0074] Taking live sports broadcasts as an example, the influencing parameters for each live broadcast may include at least one of the following parameters: schedule parameters; event parameters; and live broadcast excitement parameters. Specifically, schedule parameters indicate the parameters corresponding to the first influencing factor associated with the schedule, event parameters indicate the parameters corresponding to the second influencing factor associated with the event, and live broadcast excitement parameters indicate the parameters corresponding to the third influencing factor associated with the i-th live broadcast.

[0075] It should be understood that the first influencing factor is the schedule-related factor, including but not limited to at least one of the following: schedule attributes and the time period to which the schedule belongs. For example, schedule attributes refer to whether the schedule belongs to the quarterly season, the playoffs, or the regular season; the time period to which the schedule belongs refers to when the schedule is held, such as the spring season or the autumn season. The second influencing factor is the event-related factor, including but not limited to at least one of the following: participating teams, commentators, and event time (such as afternoon or evening sessions). Among these, participating teams can be further subdivided into various influencing factors such as the overall strength of the team, the strength of a particular player, and the team's popularity; commentators can also be subdivided into various influencing factors such as the commentator's ability and popularity; and live broadcast duration can be subdivided into various influencing factors such as total duration, commentary duration for each team / participant, match point commentary duration, and team introduction duration. The third influencing factor is the live broadcast-related factor, which can be understood as the contribution of the participating teams in the i-th event to the live broadcast of the i-th event, or as the actual performance of the participating teams in the i-th event. For example, the chat data will fluctuate based on the performance of the first and / or second teams in the i-th match. The third influencing factor can be quantified as the match metrics for both teams. For multiplayer competitive matches, the third influencing factor includes, but is not limited to, at least one of the following: total kills, total wards cleared, total Barons, total dragons, and total damage.

[0076] In some embodiments, different quantitative indicators will lead to differences in the determined influencing parameters. Referring to the foregoing, different influencing factors have different effects on the live broadcast of the event. For example, the effects of the first and second influencing factors are discrete, which will result in the schedule parameters and event parameters being discrete variables; the effect of the third influencing factor is continuous, which will result in the live broadcast excitement level parameter being continuous.

[0077] In some embodiments, for discrete variables, the corresponding influencing factors can be determined through a multiple linear regression model. These influencing factors can be understood as the influencing factors corresponding to the first influencing factor and / or the second influencing factor. Subsequently, based on the bullet screen data, multiple influencing factors can be referenced for numerical transformation to obtain the corresponding schedule parameters and / or event parameters.

[0078] In other embodiments, for continuous variables, the match data of different teams can be normalized based on the bullet screen data to obtain match data weights. These weights can be understood as the weights corresponding to the third influencing factor. Subsequently, data processing and correction are performed based on the match data weights and the match data to obtain the corresponding live broadcast excitement parameters.

[0079] It should be noted that the first, second, and third influencing factors can be categorized according to the actual needs of different schedules, events, and live broadcasts, and this application does not impose any limitations on this. Furthermore, the examples above are merely illustrative and do not limit this application; therefore, any method of obtaining the factor importance parameters of quantitative indicators for schedules, events, and live broadcasts through bullet screen data is within the scope of protection of this application and will not be elaborated further.

[0080] Step 106: Determine the playback influence parameters associated with the i-th video based on at least one influence parameter.

[0081] Schematic, the playback impact parameter is used to indicate the factor importance parameter corresponding to the non-quantitative indicator of the playback of the i-th video. In some embodiments, taking a live sports event as an example, the playback impact parameter can also be expressed as the live broadcast impact parameter.

[0082] For a video, there are quantified and non-quantified metrics. Quantified metrics can be quantified to determine their impact on playback. If we want to perform data analysis on a video, we can divide it into quantified and non-quantified metrics. Since the impact of quantified metrics on video playback has already been determined in step 104, and referring to the bullet screen data, the impact of non-quantified metrics can be determined. Here, quantified metrics can be understood as directly quantifiable metrics, and non-quantified metrics can be understood as non-quantifiable metrics.

[0083] Non-quantifiable metrics refer to those related to the i-th video that cannot be quantified. Taking live esports events as an example, non-quantifiable metrics include, but are not limited to, live broadcast incidents (such as equipment malfunctions, personal situations of participating players), and trending events of the day (such as last-minute announcements of important policy adjustments, star players participating, matches between two strong teams, and special schedules such as the esports Spring Festival Gala). It should be understood that these metrics still affect the bullet screen data, and these metrics cannot be quantified. To ensure the smooth operation of the live esports broadcast, the event organizers still need to pay attention to non-quantifiable metrics.

[0084] In some embodiments, bullet screen data is influenced by both quantitative and non-quantitative metrics.

[0085] Based on this, after determining the impact of the quantitative indicators on the i-th video according to step 104, the impact of the non-quantitative indicators on the i-th video (i.e., playback impact parameters) can be determined by referring to the bullet screen data; then, the anomalies of the i-th video can be judged and classified according to the playback impact parameters.

[0086] Step 108: Based on the n playback impact parameters corresponding to each of the n videos, predict the degree of abnormal impact of non-quantitative indicators on future video playback, and determine the playback strategy for future video playback.

[0087] According to step 106, the playback impact parameters of the i-th video can be obtained, thereby determining the impact of the non-quantitative metrics of that video. Similarly, for n videos belonging to the same series, n-1 playback impact parameters corresponding to the remaining n-1 videos (excluding the i-th video) can be obtained. Based on this, the impact of the non-quantitative metrics of the n-1 videos can be determined.

[0088] To make the data processing results more reliable, n playback impact parameters will be referenced to predict the degree of abnormality of non-quantitative indicators on future video playback, thereby providing a basis for adjusting future video playback strategies.

[0089] Optionally, the playback strategy for future video playback may include at least one of the following:

[0090] Determine the playback method for future video playback;

[0091] Determine the real-time monitoring method for future video playback;

[0092] Determine the debugging method for the video playback device, which includes at least audio and video equipment and network support equipment.

[0093] The playback method includes, but is not limited to, at least one of the following: the playback time period and playback duration of the future video. For example, by referring to n playback influence parameters and analyzing the non-quantitative indicators present in the n videos, if it is determined that the playback method has a significant impact on the playback influence parameters, then the playback method for the future video can be adjusted. It should be understood that the playback strategy for the future video can be determined solely based on non-quantitative indicators, or it can be determined based on both quantitative and non-quantitative indicators; this application does not limit this.

[0094] The illustrative "real-time monitoring method" indicates a way to monitor video playback in real time. The monitored content includes, but is not limited to, data such as bullet comments (danmaku) and the number of online users, such as changes in bullet comment data resulting from changes in quantifiable metrics. For example, by referencing n playback-influencing parameters and analyzing the non-quantifiable metrics present in n video playbacks, if a certain non-quantifiable metric is determined to have a significant impact on future video playback, then during future video playback, the focus will be on monitoring whether this non-quantifiable metric will appear, and the playback strategy will be adjusted promptly if its appearance is predicted.

[0095] Before playing a video, the video playback device needs to be tested and adjusted. If, based on n playback-influencing parameters, it's determined that network speed has a significant impact on future video playback, the testing methods for the network support equipment can be adjusted, with a focus on network speed testing. Alternatively, if analysis of non-quantifiable indicators in the n videos reveals that the quality of the audio / video equipment used for playback has a significant impact on the playback parameters, the testing methods for the video playback device can be adjusted to optimize its parameters.

[0096] Taking live sports broadcasts as an example, after obtaining n playback-influencing parameters, it is possible to review the performance of historical broadcast data (including but not limited to various data such as bullet screen data, match data, and schedule data). Based on this, event organizers can optimize the arrangement of future matches and adjust the broadcast schedule to maximize the utilization of historical broadcast data.

[0097] Optionally, the data processing method provided in this application embodiment further includes: establishing a grading mechanism for non-quantitative indicators based on the n playback impact parameters corresponding to the n events.

[0098] Referring to the foregoing, after obtaining n playback impact parameters, historical live broadcast data can be reviewed. Based on this, a grading mechanism for non-quantifiable indicators can be established to reduce their impact on future live broadcasts. For example, if a non-quantifiable indicator is a live broadcast accident, a grading mechanism for live broadcast accidents can be established to quantify their impact.

[0099] For each non-quantitative indicator, one or more grading mechanisms can be established. For example, a non-quantitative indicator is a live streaming incident, which can be divided into several types, and a grading mechanism can be established for each type.

[0100] In summary, the data processing method provided in this application presents a novel approach. By processing the bullet comment data of n videos within the same series, it is possible to determine the playback impact parameters associated with each video. Based on these parameters, the impact of non-quantitative indicators on future video playback can be predicted, thereby determining the playback strategy for future videos. This allows video organizers to adjust future video playback based on the bullet comment data.

[0101] Referring to the foregoing, taking live sports broadcasts as an example, n videos are live broadcasts of n matches belonging to the same schedule. A schedule includes at least one match, and each match corresponds to one live broadcast. Two teams compete in each match. Based on this, the data processing method provided in this application embodiment can be used to process the bullet screen data of live sports broadcasts.

[0102] Figure 4 A flowchart of a data processing method provided in an exemplary embodiment of this application is shown. The method is composed of... Figure 1 The data processing device 120 in the middle executes the method, which includes the following steps:

[0103] Step 202: Obtain the live chat data for n matches, where the n matches belong to the same schedule.

[0104] Indicatively, a schedule includes at least one event, one event corresponds to one live broadcast, and two teams compete in one event.

[0105] In this context, a tournament refers to a competition event consisting of two different teams. It can be understood as a schedule (such as a quarterly tournament or regular season) containing at least two teams, with at least one tournament consisting of two of these teams. Furthermore, when broadcasting this schedule live, each tournament can be streamed live, generating different chat data for each live stream.

[0106] For illustrative purposes, when n matches of the same schedule are broadcast live on a certain live streaming platform, the bullet screen data refers to the data information associated with the bullet screens displayed on the display interface of each match during the live broadcast.

[0107] The bullet screen data can be result-type data such as average bullet screen count, average bullet screen count per minute, and total bullet screen count, or it can be the text data carried in the bullet screen. For illustration, the average bullet screen count indicates the number of bullet screens generated by users watching the live stream per unit of time; the average bullet screen count per minute indicates the average number of bullet screens generated by users watching the live stream per minute, and can be considered a type of average bullet screen count; the total bullet screen count indicates the total number of bullet screens generated by users watching the live stream during this live stream session.

[0108] The acquisition of bullet screen data can be found in the aforementioned content.

[0109] It should be noted that the bullet screen data involved in this application can be raw data or calculated result data. If raw data is used, the data processing method provided in this application embodiment will require additional processing of the raw data.

[0110] Furthermore, subsequent data processing will be based on all the bullet screen data corresponding to the n matches. Referring to step 202, it can be understood that the data processing method provided in this application embodiment will be based on all bullet screen data belonging to the same match period, using that match period as the basis for data processing.

[0111] Step 204: Based on the bullet screen data, determine at least one influencing parameter for the live broadcast of the i-th match corresponding to the i-th match in the n matches.

[0112] Indicatively, the influence parameter is used to indicate the factor importance parameter corresponding to the quantitative indicator of the i-th live event, where n and i are positive integers greater than zero, and n is not less than i.

[0113] Among them, quantitative indicators refer to quantifiable metrics related to the live broadcast of the i-th match, including but not limited to metrics associated with this match, metrics associated with the match's schedule, and metrics involved in broadcasting the match. For example, quantitative indicators include but are not limited to at least one of the following: participating teams, match commentators, live broadcast duration, schedule attributes, and time period of the schedule.

[0114] It should be understood that all of the above indicators affect the bullet screen data. For example, if a participating team is more popular, it will lead to an increase in the number of bullet screen comments sent by users, and most of these comments will be positive and friendly. On the other hand, if the commentators have weaker communication skills, fewer people will watch the live stream, resulting in a decrease in the number of bullet screen comments.

[0115] Based on this, after obtaining the bullet screen data, and after quantifying it according to the above indicators, we can obtain the factor importance parameters corresponding to different quantitative indicators, that is, we can obtain different influence parameters.

[0116] In some embodiments, the influencing parameters include at least one of the following parameters:

[0117] • Race schedule parameters;

[0118] Among them, the schedule parameter is used to indicate the parameter corresponding to the first influencing factor associated with the schedule, which can be represented by MP. 赛程 express.

[0119] For example, the first influencing factor includes, but is not limited to, at least one of the following factors: schedule attributes and the time period to which the schedule belongs. For example, schedule attributes refer to whether the schedule belongs to the quarterly season, the playoffs, or the regular season; the time period to which the schedule belongs refers to when the schedule is held, such as the spring season or the autumn season.

[0120] It should be understood that this primary influencing factor can be quantified, and that it is a discrete influencing factor. Therefore, the schedule parameter is a discrete variable, or a 0 / 1 variable.

[0121] • Event parameters;

[0122] Among them, the event parameters are used to indicate the parameters corresponding to the second influencing factor associated with the event, and can be represented by MP. 赛事 express.

[0123] For example, the second influencing factor includes, but is not limited to, at least one of the following factors: participating teams, commentary, and match time (e.g., afternoon or evening sessions). Among these, participating teams can be further subdivided into factors such as overall team strength, the strength of individual team members, and the team's popularity; commentary can be subdivided into factors such as commentator ability and commentator popularity; and livestream duration can be subdivided into factors such as total duration, commentary duration for each team / participant, match point commentary duration, and team introduction duration.

[0124] Similar to the first influencing factor, the second influencing factor can also be quantified and is a discrete influencing factor. Therefore, the event parameters are also discrete variables (0 / 1 variables).

[0125] • Parameters for the quality of the live stream;

[0126] Among them, the live broadcast excitement level parameter is used to indicate the parameter corresponding to the third influencing factor associated with the live broadcast of the i-th event, which can be represented by MP. 直播精彩程度 express.

[0127] For example, the third influencing factor can be understood as the contribution of the participating teams in the i-th match to the live broadcast of the i-th match, or it can be understood as the actual performance of the participating teams in the i-th match. For instance, based on the performance of the first and / or second teams included in the i-th match, the chat data will fluctuate. The third influencing factor can be quantified as the match metrics of the two teams. For example, in multiplayer competitive matches, the third influencing factor includes, but is not limited to, at least one of the following: total kills, total wards cleared, total Barons, total dragons, and total damage.

[0128] It should be understood that the third influencing factor can also be quantified, but since the impact of the above-mentioned competition indicators on the live broadcast of the event is continuous, the parameter of the live broadcast's excitement level will be a continuous variable.

[0129] The determination of the corresponding influencing parameters also differs depending on whether the variables are discrete or continuous.

[0130] In some embodiments, for discrete variables, the corresponding influencing factors can be determined through a multiple linear regression model. These influencing factors can be understood as the influencing factors corresponding to the first influencing factor and / or the second influencing factor. Subsequently, based on the bullet screen data, multiple influencing factors can be referenced for numerical transformation to obtain the corresponding schedule parameters and / or event parameters.

[0131] In other embodiments, for continuous variables, the match data of different teams can be normalized based on the bullet screen data to obtain match data weights. These weights can be understood as the weights corresponding to the third influencing factor. Subsequently, data processing and correction are performed based on the match data weights and the match data to obtain the corresponding live broadcast excitement parameters.

[0132] It should be noted that the first, second, and third influencing factors can be categorized according to the actual needs of different schedules, events, and live broadcasts, and this application does not impose any limitations on this. Furthermore, the examples above are merely illustrative and do not limit this application; therefore, any method of obtaining the factor importance parameters of quantitative indicators for schedules, events, and live broadcasts through bullet screen data is within the scope of protection of this application and will not be elaborated further.

[0133] Step 206: Determine the playback impact parameters associated with the live broadcast of the i-th event based on at least one impact parameter.

[0134] Indicatively, the playback impact parameter is used to indicate the factor importance parameter corresponding to the non-quantitative indicator of the i-th live event.

[0135] For a live sports event, there are quantifiable and non-quantifiable indicators. Quantifiable indicators can be quantified to determine their impact on the live event. If we want to perform data analysis on a live sports event, we can divide it into quantifiable and non-quantifiable indicators. Since the impact of quantifiable indicators on the live event has already been determined in step 204, and referring to the chat data, the impact of non-quantifiable indicators can be determined. Quantifiable indicators can be understood as indicators that can be directly quantified, while non-quantifiable indicators can be understood as indicators that cannot be directly quantified.

[0136] Non-quantifiable indicators refer to those related to the live broadcast of the i-th match that cannot be quantified, including but not limited to live broadcast incidents (such as equipment failures, personal situations of participating players), and trending events of the day (such as last-minute announcements of important policy adjustments, the participation of star players, matches between two strong teams, and special schedules such as the e-sports Spring Festival Gala). It should be understood that these indicators still affect the bullet screen data, and these indicators cannot be quantified. To ensure the smooth broadcast of the matches, the event organizers still need to pay attention to non-quantifiable indicators.

[0137] In some embodiments, bullet screen data is influenced by both quantitative and non-quantitative metrics. Taking the average number of bullet screens per minute as an example, their interrelationship can be expressed as follows:

[0138] Average number of bullet comments per minute = MP 赛程 *MP 赛事 *MP 比赛精彩程度 *MP 直播执行 *Error.

[0139] Among them, MP 直播执行 It can be considered as a playback-affecting parameter, and Error can be considered as a comprehensive error term.

[0140] Based on this, after determining the impact of quantitative indicators on the live broadcast of the i-th event according to step 204, the impact of non-quantitative indicators on the live broadcast of the i-th event (i.e., playback impact parameters) can be determined by referring to the barrage data; then, the anomalies of the live broadcast of the i-th event can be judged and classified according to the playback impact parameters.

[0141] Step 208: Based on the n playback impact parameters corresponding to each of the n events, predict the degree of abnormal impact of non-quantitative indicators on future event live broadcasts, and determine the playback strategy for future event live broadcasts.

[0142] According to step 206, the playback impact parameters of the i-th match live stream can be obtained, thus determining the impact of non-quantitative indicators of the match live stream. Similarly, for n matches belonging to the same schedule, the playback impact parameters of the remaining n-1 matches (excluding the i-th match) can be obtained. Based on this, the impact of non-quantitative indicators of the n-1 match live streams can be determined.

[0143] To make the data processing results more reliable, we will refer to n playback impact parameters to predict the degree of impact of non-quantitative indicators on the abnormality of future live broadcasts, thereby providing a basis for adjusting the playback strategy of future live broadcasts.

[0144] Optionally, determine the broadcast strategy for future live events, including at least one of the following:

[0145] Determine the live streaming method for future events;

[0146] Determine the real-time monitoring method for future live broadcasts of events;

[0147] Determine the debugging method for the live streaming equipment, which includes at least audio and video equipment and network support equipment.

[0148] The live streaming methods include, but are not limited to, at least one of the following: the broadcast time period, broadcast duration, and selection of commentators for future live events. For example, by referring to n playback impact parameters and analyzing the non-quantitative indicators present in n live events, if it is determined that the live streaming method has a significant impact on the playback impact parameters, then the live streaming method for future live events can be adjusted. It should be understood that the playback strategy for future live events can be determined solely based on non-quantitative indicators, or it can be determined based on a combination of quantitative and non-quantitative indicators; this application does not limit this.

[0149] The illustrative "real-time monitoring method" indicates how to monitor live streams in real time. Monitoring content includes, but is not limited to, bullet screen data and the number of online viewers, such as changes in bullet screen data resulting from changes in quantifiable metrics. For example, considering n playback impact parameters and analyzing non-quantifiable metrics present in n live events, if a particular non-quantifiable metric is determined to have a significant impact on future live events, then during the playback of future live events, the focus will be on monitoring whether this non-quantifiable metric will appear, and the playback strategy will be adjusted promptly if its appearance is predicted.

[0150] Before conducting a live broadcast, the live broadcast equipment needs to be tested and adjusted. If, based on n playback impact parameters, it is determined that network speed has a significant impact on the future live broadcast of the event, the testing methods for the network support equipment can be adjusted, with a focus on network speed testing. Alternatively, if, after analyzing the non-quantifiable indicators present in the n live broadcasts of the event, it is determined that the quality of the audio and video equipment used for the broadcast has a significant impact on the playback impact parameters, the testing methods for the live broadcast equipment for the future event broadcast can be adjusted to optimize the equipment parameters.

[0151] In other embodiments, after obtaining n playback impact parameters, the performance of historical live streaming data (including but not limited to various data such as bullet screen data, match data, and schedule data) can be reviewed. Based on this, the event organizer can optimize the arrangement of future matches and adjust the arrangement of future live broadcasts to maximize the utilization of historical live streaming data.

[0152] Optionally, the data processing method provided in this application embodiment further includes: establishing a grading mechanism for non-quantitative indicators based on the n playback impact parameters corresponding to the n events.

[0153] Referring to the foregoing, after obtaining n playback impact parameters, historical live broadcast data can be reviewed. Based on this, a grading mechanism for non-quantifiable indicators can be established to reduce their impact on future live broadcasts. For example, if a non-quantifiable indicator is a live broadcast accident, a grading mechanism for live broadcast accidents can be established to quantify their impact.

[0154] For each non-quantitative indicator, one or more grading mechanisms can be established. For example, a non-quantitative indicator is a live streaming incident, which can be divided into several types, and a grading mechanism can be established for each type.

[0155] In summary, the data processing method provided in this application presents a novel approach. By processing the live chat data of n matches within the same schedule, it is possible to determine the playback impact parameters associated with each match's live stream. Based on these playback impact parameters, the impact of non-quantitative indicators on future live streams can be predicted, thereby determining the playback strategy for future live streams. This allows event organizers to adjust future live streams based on the chat data.

[0156] refer to Figure 4 , Figure 5 The flowchart illustrates a data processing method provided in an exemplary embodiment of this application, and gives an optional method for determining the influence parameters and playback influence parameters, which can be specifically combined with... Figure 6 The flowchart is shown. Step 204 can be implemented as steps 2041 and 2051, or as steps 2041, 2052, and 2053. Steps 2051 and 2052 can be executed selectively, simultaneously, or not simultaneously; this application does not limit this. Meanwhile, step 206 is implemented as steps 2061 and 2062. Specifically, as follows:

[0157] Step 2041: Determine the average number of bullet comments based on the bullet comment data.

[0158] For illustrative purposes, the average number of bullet comments (danmaku) indicates the number of bullet comments generated by users watching a live stream per unit of time. Referring to the foregoing, bullet comment data can be raw data obtained based on bullet comment capture technology, which can then be processed to obtain the average number of bullet comments.

[0159] The unit of time can be set according to actual needs, such as per match, per hour, or per minute. In some embodiments, the average number of bullet comments per minute can be considered as the average number of bullet comments per minute for subsequent data analysis. The average number of bullet comments per minute is a type of average bullet comment volume, used to indicate the average number of bullet comments generated per minute by users watching the live stream.

[0160] The following examples will all be described using the average number of bullet comments per minute as the average number of bullet comments per minute.

[0161] Step 2051: Determine the schedule parameters and / or event parameters based on the average number of bullet comments.

[0162] Referring to the foregoing, since both the schedule parameters and the event parameters are discrete variables, their determination methods are similar.

[0163] For example, the corresponding influencing factors can be determined using a multiple linear regression model. These influencing factors can be understood as the influencing factors corresponding to the first and / or second influencing factors. Subsequently, based on the bullet screen data, multiple influencing factors can be referenced for numerical conversion to obtain the corresponding schedule parameters and / or event parameters. The relevant descriptions of the first and / or second influencing factors can be found in the aforementioned content.

[0164] Optionally, step 2051 can be implemented as follows:

[0165] Based on the average number of bullet comments and a multiple linear regression model, determine the influencing factors of the schedule and / or the event.

[0166] The competition schedule parameters are determined by numerical conversion based on the competition's impact factor and average number of bullet comments; and / or, the competition parameters are determined by numerical conversion based on the competition's impact factor and average number of bullet comments.

[0167] The multiple linear regression model is used to describe the relationship between bullet screen data and the first and / or second influencing factors. The first and / or second influencing factors can be represented by match data. It should be understood that since the match schedule parameters and / or event parameters are the result of the combined influence of multiple factors, the multiple linear regression model can be used for breakdown and quantification.

[0168] For example, the impact factor of the competition schedule is determined based on the multiple linear relationship between the average number of bullet comments per minute and the first influencing factor; similarly, the impact factor of the event is determined based on the multiple linear relationship between the average number of bullet comments per minute and the second influencing factor. Here, the multiple linear relationship is the relationship represented by a multiple linear regression model.

[0169] In some embodiments, a dependent variable in a multiple linear regression model is determined by multiple independent variables. In this model, an observation (dependent variable) is a one-dimensional scalar, and the independent variables (X) are high-dimensional vectors. The data space of a multiple linear regression model can be represented as (y1, y2, ..., y...). q ), where each y k =β0+β1x1+…+β p x p +ε, where x is the independent variable.

[0170] refer to Figure 6 For a multiple linear regression model, the first influencing factor / second influencing factor is the independent variable x, the average number of bullet comments is the dependent variable y, and the coefficient β obtained by solving the multiple linear regression model is the influencing factor of the schedule / event.

[0171] Optionally, the impact factors of the schedule and / or the event can be implemented in the following ways:

[0172] Based on the multiple linear relationship between the average number of bullet comments and the primary influencing factor, the multiple linear regression equation of the competition schedule is determined to identify the influencing factors of the competition schedule.

[0173] Based on the multiple linear relationship between the average number of bullet comments and the second influencing factor, the multiple linear regression equation of the event is determined to identify the event's influencing factors.

[0174] Referring to the foregoing, matrices can be used to express multivariate linear relationships, as follows:

[0175] y = β0 + β1x1 + ... + β p x p +∈;

[0176] E(y) = β0 + β1x1 + ... + β p x p ;

[0177]

[0178] For the given data points, the problem simplifies to minimizing the following objective function:

[0179]

[0180] Referring to the objective function above, taking the derivative of each term yields:

[0181] ∑(y i -β0-β1x i1 -…-β p x ip ) = 0;

[0182] ∑(y i -β0-β1x i1 -…-β p x ip )x i1 =0;

[0183] …;

[0184] ∑(y i -β0-β1x i1 -…-β p x ip )x ip =0.

[0185] In some embodiments, the matrix derivative method can also be used to obtain the following matrix solution:

[0186]

[0187]

[0188]

[0189]

[0190] When there is no strong correlation among multiple influencing factors, the columns of matrix X are full rank, therefore X T X is an invertible matrix, and the solution can be achieved through matrix inversion operations. Through the above calculations, the correlation coefficients used as additive operators can be obtained. In some embodiments, the calculated correlation coefficients need to undergo significance testing, i.e., determining the p-value (an indicator of statistical significance of variables) for all coefficients, and the p-value needs to pass a threshold test of 0.05 (indicating that its global effect is significant). Subsequently, through R... 2 The goodness of fit is measured by an index (R0) that reflects the degree of fit between the estimated and actual values. 2 This expresses the ratio of the regression sum of squares to the total sum of squares, reflecting the proportion of the dependent variable that can be explained by the independent variable:

[0191]

[0192] It should be noted that, in the embodiments of this application, after calculating the data of a certain race period, the obtained R is... 2 The accuracy of the multiple linear regression model is 0.947, which confirms its high fitting accuracy.

[0193] refer to Figure 6 After determining the influencing factors of the schedule / event, the schedule parameter MP can be determined through numerical transformation. 赛程 and / or event parameters MP 赛事 .

[0194] In some embodiments, the numerical transformation is based on the following variables: the influence factors of the schedule and / or the event; the average value of the influence factors of the schedule and / or the event; and the average number of bullet comments per minute. Optionally, the numerical transformation can be performed in the following ways:

[0195] Based on the sum of the influence factors of the competition schedule and the average of the influence factors of the competition schedule, and the ratio of the average number of bullet comments, the influence parameters corresponding to each influence factor of the competition schedule are calculated; the influence parameters corresponding to each influence factor of the competition schedule are multiplied together to obtain the competition schedule parameters.

[0196] And / or, based on the sum of the event's impact factors and the average of the event's impact factors, and the ratio of the average number of bullet comments, calculate the impact parameter corresponding to each impact factor of the event; multiply the impact parameters corresponding to each impact factor of the event together to obtain the event parameters.

[0197] For example, MP 赛程 / MP 赛事 It can be calculated using the following formula:

[0198]

[0199]

[0200] Among them, MP 赛程or赛事某因子 For each influencing factor of the schedule / event, the corresponding influence parameter is called coef. 赛程or赛事某因子 The Base is the influencing factor of the schedule / event (i.e., the coefficient β calculated through a multiple linear regression model). Coef赛程or赛事某因子 This represents the average impact factor of the schedule / event, while DependetVariableBase represents the average number of bullet comments.

[0201] Referring to the above calculation formula, after calculating the impact parameter corresponding to each impact factor, multiple impact factors can be multiplied together, and their product can be determined as the corresponding schedule parameter / event parameter.

[0202] The multiple linear regression model is used to describe the relationship between bullet screen data and the first and / or second influencing factors. Through the above calculation process, the quantitative impact of the schedule and / or event on the live broadcast of the i-th event can be determined, thus providing a basis for the determination of subsequent non-quantitative indicators.

[0203] Step 2052: Determine the team influence parameters for the i-th match live stream based on the average number of bullet comments.

[0204] Referring to the foregoing, since the live stream excitement level parameter is a discrete variable, the match data of different teams can be normalized based on the bullet screen data to obtain the match data weights. These weights can be understood as the weights corresponding to the third influencing factor. Subsequently, based on the match data weights and the match data, data processing and correction are performed to obtain the corresponding live stream excitement level parameter.

[0205] In some embodiments, team influence parameters can be calculated using a multiple linear regression model, and this calculation process can be performed simultaneously during the calculation of schedule parameters / event parameters. Specifically, based on the average number of bullet comments corresponding to at least one event participated in by each team throughout the entire schedule, the team influence parameters in the live broadcast corresponding to that event are determined using a multiple linear regression model.

[0206] Step 2053: Determine the live broadcast excitement level parameter based on the team influence parameter and the match data of the i-th match.

[0207] The match data for the i-th event is used to indicate the data generated by the first and / or second teams in the i-th event that are related to the match result. Taking multiplayer competitive matches as an example, the match data may include at least one of the following: total kills, total wards cleared, total Barons, total dragons, and total damage.

[0208] refer to Figure 6 The calculation of the live broadcast excitement coefficient involves multiple steps, primarily based on the average performance of the two participating teams in the i-th match. For example, based on each team's match data in the i-th match and their average match data throughout the entire tournament (i.e., the average of the match data across the entire tournament), the excitement level of that team based on their average performance is determined, and this excitement level is reflected by the excitement coefficient. Subsequently, the excitement coefficient is adjusted using a team influence parameter to obtain an excitement conversion ratio. Finally, by converting the excitement conversion ratio, the live broadcast excitement parameter for the i-th match is calculated.

[0209] Optionally, step 2053 can be implemented as follows:

[0210] Based on the match data of the i-th match, determine the excitement level coefficient of the live broadcast of the i-th match;

[0211] The excitement level coefficient is corrected based on the team influence parameters to obtain the excitement level conversion ratio;

[0212] The live stream's excitement level parameter is obtained by converting the excitement level conversion ratio based on the logit nonlinear transformation.

[0213] For example, the live broadcast quality parameter for the i-th match can be divided into the following three steps:

[0214] Step 1: Determine the excitement level coefficient of the i-th match live broadcast.

[0215] Optionally, the correlation coefficient between the match data and the average number of bullet comments is normalized to obtain the match data weight corresponding to the match data; an intermediate parameter is calculated based on the ratio of the match data to the average match data; and the excitement level coefficient is determined based on the product of the match data weight and the intermediate parameter.

[0216] In some embodiments, the aforementioned competition data can be considered a specific manifestation of in-game competition metrics. Correlation analysis verifies a significant positive correlation between the bullet screen data and in-game competition metrics, indicating that these in-game factors reflect the excitement level of the competition and ultimately affect the live stream's excitement level parameters. Furthermore, based on descriptive data statistics associated with multiple competition periods, significant differences were found in the baseline performance values ​​of different participating teams at different times.

[0217] Therefore, it is necessary to normalize the correlation coefficient between the match data and the average number of bullet comments to obtain the corresponding match data weights. These weights can be represented by the Weight function. 赛内指标i This is represented by an index. Based on this, the excitement level coefficient can be determined according to the weight of the match data. The excitement level coefficient can be represented by an index. 比赛精彩程度 express.

[0218] For example, the excitement level coefficient can be calculated using the following formula:

[0219]

[0220] Among them, the index i = 1, 2, 3, 4, 5 is used to indicate the total number of kills, total number of wards cleared, total number of Barons, total number of dragons, and total damage; the average of index i for team k during the season j is used to indicate the average value of the match data corresponding to the index.

[0221] Step 2: Determine the conversion rate of the level of excitement.

[0222] Optionally, based on the team influence parameter, all teams participating in the competition are classified to obtain the calibration index of the first and / or second team in the i-th match; the excitement conversion ratio is obtained based on the ratio of the excitement coefficient to the calibration index.

[0223] The calculation of the team impact parameters can be found in the aforementioned content and will not be repeated here.

[0224] Among the participating teams in a particular tournament, there are differences in their abilities, which can be used to distinguish between strong, medium, and weak teams. Furthermore, the impact of different teams on the excitement of the match varies. For example, a strong team performing poorly will result in a greater loss of viewers compared to a weaker team, and conversely, a strong team performing well will generate more viewers than a weaker team.

[0225] refer to Figure 6 After calculating the excitement level coefficient, it can be corrected by the team influence parameter to obtain the excitement level conversion ratio, so as to correct the impact of different participating teams on the live broadcast of the event.

[0226] Using MP with team impact parameters 队伍 For example, MP is calculated based on the aforementioned content. 队伍 It should be a number around 1. Among them, MP 队伍 >1 represents above average, MP 队伍 <1 indicates below average; the larger the value, the stronger the team's ability.

[0227] Taking a certain tournament with 17 teams as an example, according to MP 队伍The teams were divided into six strong teams, four medium-strength teams, and seven weak teams. The excitement level coefficient for each team can be calculated using the following formula:

[0228]

[0229]

[0230]

[0231] That is, the strong team's adjustment index is 1 (i.e., no adjustment), and only the weak and medium teams are adjusted.

[0232] Subsequently, in the i-th match, the corresponding excitement level conversion ratio for the participating teams (team one and / or team two) can be calculated using the following formula:

[0233]

[0234] Among them, Adj 对应校准指数 Used to indicate the calibration index corresponding to the first team and / or the second team, which can be a strong team, a medium team, or a weak team.

[0235] Step 3: Determine the parameters for assessing the quality of the live stream.

[0236] It should be understood that, given the limited total potential audience, the viewing metrics will not linearly approach 0 or infinity. This can be interpreted as the relationship between the live stream's entertainment value and the average number of bullet comments being non-linear. The relationship can be found by referring to... Figure 7 .

[0237] Exemplary Reference Figure 6 The live stream's excitement level can be further transformed using a logit nonlinear transformation to obtain the live stream excitement level parameter, as shown in the following formula:

[0238]

[0239] Since the live broadcast excitement parameter is a discrete variable, the weight of the third influencing factor is first determined by normalizing the match data of different teams. Then, the live broadcast excitement parameter is adjusted and corrected according to the weight and match data to determine the quantitative impact of the live broadcast process on the i-th match. In this way, non-quantitative indicators are determined by combining the schedule and / or the impact of the match.

[0240] Step 2061: Determine the average predicted value of bullet comments based on at least one influencing parameter.

[0241] The illustrative average bullet comment prediction value is used to indicate the predicted average number of bullet comments for a given schedule of n events. (Reference) Figure 6 Based on the aforementioned steps, the schedule parameters, event parameters, and live broadcast excitement parameters can be calculated. Subsequently, the average barrage prediction value can be calculated based on the aforementioned multiple influencing parameters.

[0242] The average predicted value of bullet comments can be represented by Average.

[0243] In some embodiments, the average predicted bullet screen value can be obtained through the following calculation process:

[0244] (1) Determine the impact parameters of the i-th live event, obtain the product of multiple impact parameters, and divide the actual number of bullet comments of the i-th live event by the aforementioned product to obtain the predicted number of bullet comments of the i-th live event. This number is the predicted number of bullet comments that should be obtained if there are no multiple impact factors.

[0245] (2) Similar to step (1), obtain the predicted number of bullet comments for the remaining n-1 live events;

[0246] (3) Based on the predicted number of bullet comments for n live events, obtain their average value and determine the average predicted bullet comment value.

[0247] It should be understood that the average predicted value of the live stream comments is a prediction of the average level of the live stream for this event based on at least one influencing parameter. When performing data analysis on the live stream of the i-th event, it is necessary to base the analysis on this predicted average level to ensure the accuracy of the data analysis.

[0248] Step 2062: Determine the playback impact parameters associated with the live broadcast of the i-th event based on the average predicted value of the bullet comments and the average number of bullet comments in the live broadcast of the i-th event.

[0249] refer to Figure 6 After determining the average predicted value of bullet comments, it is necessary to introduce the average number of bullet comments for the i-th live event to determine the playback impact parameters associated with the i-th live event. For example, the playback impact parameters associated with the i-th live event can be calculated using the following formula:

[0250]

[0251] It should be understood that, given that the live broadcast of the event may be affected by other unconsidered factors, and that the calculation of the aforementioned multiple influencing parameters may contain errors, the playback impact parameters calculated here will be approximate values, hence the use of an approximate equality sign.

[0252] Based on this, the playback impact parameters associated with the live broadcast of the i-th event will be obtained.

[0253] In summary, the data analysis method provided in this application provides calculation methods for multiple influencing parameters, and also provides specific calculation methods for playback influencing parameters. By calculating the average number of bullet comments, multiple influencing parameters of live sports events can be accurately calculated to determine the degree of influence of different quantitative indicators on live sports events, thereby facilitating the prediction of subsequent non-quantitative indicators.

[0254] It should be understood that the various calculation methods given above are merely illustrative examples and do not limit this application. Furthermore, it should be understood that any method that calculates playback impact parameters based on the average number of bullet comments is within the scope of protection of this application and will not be elaborated further.

[0255] Figure 8 A flowchart of a data processing method provided in an exemplary embodiment of this application is shown.

[0256] Among them, the match data can be referred to the above content; the match schedule data is used to indicate the data related to the match schedule, which can be understood as the data corresponding to the first influencing factor mentioned above; the viewing indicators include the aforementioned bullet screen data, and may also include the number of concurrent users (ACU).

[0257] In some embodiments, taking the average number of bullet comments per minute as an example, the correlation between the average number of bullet comments per minute and the average number of bullet comments (ACU) can be verified to confirm their synergistic relationship and representativeness. For example, this can be verified using the correlation coefficient, a statistical indicator reflecting the degree of relationship between variables. The correlation coefficient ranges from 1 to -1. A correlation coefficient of 1 indicates a perfectly linear correlation between the two variables; a correlation coefficient of -1 indicates a perfectly negative correlation; and a correlation coefficient of 0 indicates no correlation. The closer the data is to 0, the weaker the correlation.

[0258] For example, the formula for calculating the correlation coefficient is as follows:

[0259]

[0260]

[0261]

[0262]

[0263] Where x and y are used to indicate the average number of bullet comments per minute and ACU, respectively, r xy The correlation coefficient used to indicate the average number of bullet comments per minute and ACU, s x and s yThese are used to indicate the standard deviation of the average number of bullet comments per minute and the average number of ACUs, respectively. xy Used to indicate the covariance of average bullet screen volume and ACU.

[0264] It should be noted that, based on the average number of bullet comments per minute and ACU across multiple competition sessions, the correlation coefficient between the average number of bullet comments per minute and ACU is determined to be 0.82, proving that there is a very strong positive correlation between the average number of bullet comments per minute and ACU. Therefore, the average number of bullet comments per minute and ACU can be used as substitute reference indicators for each other.

[0265] refer to Figure 8 Based on match data, schedule data, and spectator metrics, comprehensive spectator data can be obtained. After anomaly identification and cleanup of the comprehensive spectator data, valid spectator data is obtained. Subsequently, different influencing parameters can be calculated based on the valid spectator data.

[0266] For example, sub-model 1 is used to determine discrete variables, namely, the schedule parameters and / or event parameters; sub-model 2 is used to determine continuous variables, namely, the parameters of the live broadcast's excitement level.

[0267] The schedule parameters and / or event parameters can be calculated using a multiple linear regression model, as detailed above. Additionally, after calculating the influencing factors of the schedule and / or event, the baseline values ​​for the entire schedule can be calculated to facilitate the calculation of sub-model 2.

[0268] For example, the base value for the entire race can be calculated using the following formula:

[0269]

[0270] Among them, Base Parameter The base value used to indicate the entire competition; the average number of bullet comments per minute (Base) can also be represented as DependetVariableBase, where Base indicates the average number of bullet comments per minute; CountofCoef 赛程or赛事因子 Used to indicate the sum of factors influencing the schedule / event.

[0271] In sub-model 2, the average performance of each team in the corresponding schedule can be understood as follows: For each team, there is at least one match to participate in throughout the entire schedule. Based on the match data of all the matches each team participates in, the average value of these matches is determined as the average performance of each team in the corresponding schedule. Based on this average value, the average level of each team can be obtained, which can be used to judge the level of performance in each match.

[0272] After determining the parameters for the schedule, the event, and the live broadcast's level of excitement, the parameters affecting playback can be determined. The specific determination process can be found in the aforementioned content and will not be repeated here.

[0273] This illustration demonstrates that by determining the n playback impact parameters for each of the n matches, the degree of abnormal impact of non-quantitative indicators on future live broadcasts can be predicted, thereby determining the playback strategy for future live broadcasts. (Reference) Figure 8 The output results include three aspects, which describe the final application scenario of the data analysis model output provided in this application embodiment, namely, the degree of abnormal impact of the quantification playback influence parameters on future live sports events.

[0274] Among them, execution performance is used to quantify the performance of live streaming, which can be determined by the average number of bullet comments per minute and / or the loss of ACU; the impact of live streaming accidents is used to quantify the impact of various live streaming accidents on the viewing indicators, which can be determined by classifying the severity of the accidents; the impact of excitement level is used to quantify the impact of the excitement level of the competition on the viewing indicators.

[0275] refer to Figure 8 This application provides a new method for quantitative analysis of indicators and gives a specific calculation process for relevant influencing factors, making the data analysis method applicable to data analysis of live broadcasts of any type of competition, and has wide applicability.

[0276] Meanwhile, based on the embodiments of this application, factors that may affect viewing metrics are modeled separately, achieving a detailed breakdown of these metric factors. In some embodiments, based on relevant data from the four major seasons (Spring Regular Season / Summer Playoffs / Autumn Regular Season / Winter Regular Season) of a multiplayer competitive game throughout the year, Figure 8 The error metric for the given data analysis model is 0.152. More than 15% of the points were caused by non-quantifiable hot topics or major live broadcast incidents. Based on this, the event organizers will pay closer attention to hot topics and major live broadcast incidents in future event broadcasts and adjust the broadcast strategies accordingly, including those related to the two non-quantifiable metrics mentioned above.

[0277] In summary, the data analysis method provided in this application presents a novel data processing approach. Specifically, by processing the live chat data of n matches within the same schedule, it is possible to determine the playback impact parameters associated with each match's live stream. Based on these playback impact parameters, the impact of non-quantitative indicators on future match live streams can be predicted, thereby determining the playback strategy for future match live streams. This allows event organizers to adjust future match live streams based on the chat data.

[0278] The following are device embodiments of this application. For details not described in detail in the device embodiments, please refer to the corresponding descriptions in the above method embodiments. They will not be repeated here.

[0279] Figure 9A schematic diagram of a data processing apparatus provided in an exemplary embodiment of this application is shown. The apparatus includes:

[0280] Module 910 is used to acquire the bullet screen data of n videos, where the n videos belong to the same series;

[0281] The processing module 920 is used to determine at least one influence parameter of the i-th video among n videos based on the bullet screen data. The influence parameter is used to indicate the factor importance parameter corresponding to the quantitative index of the i-th video.

[0282] The processing module 920 is further configured to determine the playback influence parameter associated with the i-th video based on at least one influence parameter, wherein the playback influence parameter is used to indicate the factor importance parameter corresponding to the non-quantitative indicator of the i-th video;

[0283] The processing module 920 is also used to predict the degree of abnormal impact of non-quantitative indicators on future video playback based on the n playback impact parameters corresponding to each video, and to determine the playback strategy for future video playback.

[0284] Where n and i are positive integers greater than zero, and n is not less than i.

[0285] Optionally, the n videos are live streams of n events belonging to the same schedule. A schedule includes at least one event, and each event corresponds to a live stream. There are two teams competing in each event. The processing module 920 is used to determine at least one influence parameter of the live stream of the i-th event corresponding to the i-th event in the n events based on the bullet screen data. Based on the at least one influence parameter, the live stream influence parameter associated with the i-th event live stream is determined.

[0286] Optionally, the influencing parameters include at least one of the following parameters: schedule parameters; event parameters; live broadcast excitement parameters; wherein, the schedule parameters are used to indicate the parameters corresponding to the first influencing factor associated with the schedule, the event parameters are used to indicate the parameters corresponding to the second influencing factor associated with the event, and the live broadcast excitement parameters are used to indicate the parameters corresponding to the third influencing factor associated with the live broadcast of the i-th event.

[0287] Optionally, the processing module 920 is used to determine the average number of bullet comments based on the bullet comment data, the average number of bullet comments being used to indicate the number of bullet comments generated by users watching the live stream per unit of time; and to determine the schedule parameters and / or event parameters based on the average number of bullet comments.

[0288] Optionally, the processing module 920 is used to determine the schedule and / or the event's influencing factors based on the average number of bullet comments and a multiple linear regression model; to perform numerical transformation based on the schedule's influencing factors and the average number of bullet comments to determine the schedule parameters; and / or to perform numerical transformation based on the event's influencing factors and the average number of bullet comments to determine the event parameters.

[0289] Optionally, the impact factors of the schedule and / or the event are determined as follows: based on the multiple linear relationship between the average number of bullet comments and the first influencing factor, the multiple linear regression equation of the schedule is determined to determine the impact factors of the schedule; based on the multiple linear relationship between the average number of bullet comments and the second influencing factor, the multiple linear regression equation of the event is determined to determine the impact factors of the event.

[0290] Optionally, numerical transformation can be achieved as follows: Calculate the influence parameter corresponding to each influence factor of the competition based on the ratio of the sum of the influence factors of the competition and the average number of influence factors to the average number of bullet comments; multiply the influence parameters corresponding to each influence factor of the competition to obtain the competition parameters; and / or, calculate the influence parameter corresponding to each influence factor of the event based on the ratio of the sum of the influence factors of the event and the average number of influence factors to the average number of bullet comments; multiply the influence parameters corresponding to each influence factor of the event to obtain the event parameters.

[0291] Optionally, the processing module 920 is used to determine the average number of bullet comments based on the bullet comment data, whereby the average number of bullet comments indicates the number of bullet comments generated by users watching the live stream per unit of time; to determine the team influence parameters for the live stream of the i-th match based on the average number of bullet comments; and to determine the live stream excitement level parameters based on the team influence parameters and the match data of the i-th match.

[0292] Optionally, the processing module 920 is used to determine the influence factors of the first team and / or the second team in the i-th live broadcast of the competition based on the average number of bullet comments and a multiple linear regression model; and to determine the team influence parameters based on the influence factors of the first team and / or the second team in the competition.

[0293] Optionally, the processing module 920 is used to determine the excitement level coefficient of the live broadcast of the i-th match based on the match data of the i-th match; to correct the excitement level coefficient based on the team influence parameter to obtain the excitement level conversion ratio; and to convert the excitement level conversion ratio based on the logit nonlinear transformation to obtain the live broadcast excitement level parameter.

[0294] Optionally, the processing module 920 is used to normalize the correlation coefficient between the match data and the average number of bullet comments to obtain the match data weight corresponding to the match data; calculate the intermediate parameter based on the ratio of the match data to the average match data; and determine the excitement coefficient based on the product of the match data weight and the intermediate parameter.

[0295] Optionally, the processing module 920 is used to classify all teams participating in the competition according to the team influence parameter, and obtain the calibration index of the first team and / or the second team in the i-th game; and obtain the excitement conversion ratio according to the ratio of the excitement coefficient to the calibration index.

[0296] Optionally, the processing module 920 is used to determine an average predicted value of bullet comments based on at least one influence parameter, the average predicted value of bullet comments indicating the predicted value of the average number of bullet comments in the schedule to which the n events belong; and to determine the playback influence parameter based on the average predicted value of bullet comments and the average number of bullet comments in the live broadcast of the i-th event.

[0297] Optionally, the processing module 920 is used to implement at least one of the following: determining the playback mode of future video playback; determining the real-time monitoring mode of future video playback; determining the debugging mode of the video playback device, wherein the video playback device includes at least audio and video devices and network support devices.

[0298] Please refer to Figure 10 This illustration shows a structural block diagram of a computer device 1000 provided in an exemplary embodiment of this application. The computer device 1000 may be a portable mobile terminal, such as a smartphone, tablet computer, MP3 player (Moving Picture Experts Group Audio Layer III), or MP4 player (Moving Picture Experts Group Audio Layer IV). The computer device 1000 may also be referred to as a user device, portable terminal, or other names.

[0299] Typically, computer device 1000 includes a processor 1001 and a memory 1002.

[0300] Processor 1001 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 1001 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 1001 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 1001 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 1001 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0301] The memory 1002 may include one or more computer-readable storage media, which may be tangible and non-transitory. The memory 1002 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 1002 are used to store at least one instruction, which is executed by the processor 1001 to implement the data analysis method provided in the embodiments of this application.

[0302] In some embodiments, the computer device 1000 may also optionally include: a peripheral device interface 1003 and at least one peripheral device. Specifically, the peripheral device includes at least one of: a radio frequency circuit 1004, a touch display screen 1005, a camera assembly 1006, an audio circuit 1007, and a power supply 1008.

[0303] Peripheral device interface 1003 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 1001 and memory 1002. In some embodiments, processor 1001, memory 1002 and peripheral device interface 1003 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 1001, memory 1002 and peripheral device interface 1003 can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0304] The radio frequency (RF) circuit 1004 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 1004 communicates with communication networks and other communication devices via electromagnetic signals. The RF circuit 1004 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. Optionally, the RF circuit 1004 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF circuit 1004 can communicate with other terminals through at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: the World Wide Web, metropolitan area networks, intranets, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks, and / or WiFi networks. In some embodiments, the RF circuit 1004 may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0305] The touch display screen 1005 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. The touch display screen 1005 also has the ability to acquire touch signals on or above its surface. These touch signals can be input as control signals to the processor 1001 for processing. The touch display screen 1005 is used to provide virtual buttons and / or a comment keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, the touch display screen 1005 may be a single screen, located on the front panel of the computer device 1000; in other embodiments, there may be at least two touch display screens, respectively located on different surfaces of the computer device 1000 or in a folded design; in still other embodiments, the touch display screen 1005 may be a flexible display screen, located on a curved or folded surface of the computer device 1000. Furthermore, the touch display screen 1005 may also be configured as a non-rectangular, irregular shape, i.e., a non-rectangular screen. The touch display screen 1005 can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).

[0306] The camera assembly 1006 is used to capture images or videos. Optionally, the camera assembly 1006 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is used for video calls or selfies, and the rear-facing camera is used for taking photos or videos. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, and a wide-angle camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, and panoramic shooting and VR (Virtual Reality) shooting by fusion of the main camera and the wide-angle camera. In some embodiments, the camera assembly 1006 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash is a combination of a warm-light flash and a cool-light flash, which can be used for light compensation at different color temperatures.

[0307] Audio circuit 1007 provides an audio interface between the user and computer device 1000. Audio circuit 1007 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting the sound waves into electrical signals that are input to processor 1001 for processing, or input to radio frequency circuit 1004 for voice communication. For stereo sound acquisition or noise reduction purposes, multiple microphones may be used, each located at a different location within computer device 1000. The microphone may also be an array microphone or an omnidirectional microphone. The speaker converts electrical signals from processor 1001 or radio frequency circuit 1004 into sound waves. The speaker may be a conventional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, audio circuit 1007 may also include a headphone jack.

[0308] Power supply 1008 is used to supply power to the various components in computer device 1000. Power supply 1008 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When power supply 1008 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged via a wired line, and a wireless rechargeable battery is a battery that is charged via a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0309] In some embodiments, the computer device 1000 further includes one or more sensors 1009. The one or more sensors 1009 include, but are not limited to, an accelerometer 1010, a gyroscope 1011, a pressure sensor 1012, an optical sensor 1013, and a proximity sensor 1014.

[0310] Accelerometer 1010 can detect the magnitude of acceleration along the three coordinate axes of a coordinate system established by computer device 1000. For example, accelerometer 1010 can be used to detect the components of gravitational acceleration along the three coordinate axes. Processor 1001 can control touchscreen 1005 to display the user interface in landscape or portrait view based on the gravitational acceleration signal acquired by accelerometer 1010. Accelerometer 1010 can also be used for games or for acquiring user motion data.

[0311] The gyroscope sensor 1011 can detect the orientation and rotation angle of the computer device 1000. The gyroscope sensor 1011 can work in conjunction with the accelerometer sensor 1010 to acquire the user's 3D movements on the computer device 1000. Based on the data acquired by the gyroscope sensor 1011, the processor 1001 can perform the following functions: motion sensing (e.g., changing the UI based on the user's tilt), image stabilization during shooting, game control, and inertial navigation.

[0312] The pressure sensor 1012 can be disposed on the side bezel of the computer device 1000 and / or on the lower layer of the touch display screen 1005. When the pressure sensor 1012 is disposed on the side bezel of the computer device 1000, it can detect the user's grip signal on the computer device 1000 and perform left / right hand recognition or quick operation based on the grip signal. When the pressure sensor 1012 is disposed on the lower layer of the touch display screen 1005, it can control operable controls on the UI interface based on the user's pressure operation on the touch display screen 1005. Operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

[0313] An optical sensor 1013 is used to collect ambient light intensity. In one embodiment, the processor 1001 can control the display brightness of the touch screen 1005 based on the ambient light intensity collected by the optical sensor 1013. Specifically, when the ambient light intensity is high, the display brightness of the touch screen 1005 is increased; when the ambient light intensity is low, the display brightness of the touch screen 1005 is decreased. In another embodiment, the processor 1001 can also dynamically adjust the shooting parameters of the camera assembly 1006 based on the ambient light intensity collected by the optical sensor 1013.

[0314] The proximity sensor 1014, also known as a distance sensor, is typically located on the front of the computer device 1000. The proximity sensor 1014 is used to detect the distance between the user and the front of the computer device 1000. In one embodiment, when the proximity sensor 1014 detects that the distance between the user and the front of the computer device 1000 is gradually decreasing, the processor 1001 controls the touchscreen display 1005 to switch from a screen-on state to a screen-off state; when the proximity sensor 1014 detects that the distance between the user and the front of the computer device 1000 is gradually increasing, the processor 1001 controls the touchscreen display 1005 to switch from a screen-off state to a screen-on state.

[0315] Those skilled in the art will understand that Figure 10 The structure shown does not constitute a limitation on the computer device 1000, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0316] This application also provides a computer device including a memory and a processor; the memory stores at least one piece of program code, which is loaded and executed by the processor to implement the data analysis method described above.

[0317] This application also provides a computer-readable storage medium storing a computer program that is executed by a processor to implement the data analysis method described above.

[0318] This application also provides a chip, which includes programmable logic circuits and / or program instructions, for implementing the data analysis method described above when the electronic device on which the chip is installed is running.

[0319] This application also provides a computer program product, which includes computer instructions stored in a computer-readable storage medium. The processor reads and executes the computer instructions from the computer-readable storage medium to implement the data analysis method described above.

[0320] It should be understood that "multiple" as used in this article refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0321] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0322] The above description is merely an optional embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A data processing method, characterized in that, The method includes: Obtain the bullet screen data of n videos, wherein the n videos belong to the same series; the n videos are live videos of n events belonging to the same schedule, a schedule includes at least one event, one event corresponds to one live broadcast, and there are two teams competing in one event; Based on the bullet screen data, at least one influence parameter is determined for the live broadcast of the i-th event corresponding to the i-th event in the n events. The influence parameter is used to indicate the factor importance parameter corresponding to the quantitative indicator for playing the live broadcast of the i-th event. Based on the at least one influence parameter, determine the live broadcast influence parameter associated with the live broadcast of the i-th event. The live broadcast influence parameter is used to indicate the factor importance parameter corresponding to the non-quantitative indicator for broadcasting the live broadcast of the i-th event. Based on the n live streaming impact parameters corresponding to the n videos, predict the degree of abnormal impact of the non-quantitative indicators on future video playback, and determine the playback strategy for future video playback; Where n and i are positive integers greater than zero, and n is not less than i.

2. The method according to claim 1, characterized in that, The influencing parameters include at least one of the following parameters: Race schedule parameters; Event parameters; Parameters for assessing the quality of a live stream; The schedule parameter indicates the parameter corresponding to the first influencing factor associated with the schedule, the event parameter indicates the parameter corresponding to the second influencing factor associated with the event, and the live broadcast excitement parameter indicates the parameter corresponding to the third influencing factor associated with the live broadcast of the i-th event.

3. The method according to claim 2, characterized in that, The step of determining at least one influencing parameter for the live broadcast of the i-th match corresponding to the i-th match in the n matches based on the bullet screen data includes: The average number of bullet comments is determined based on the bullet comment data. The average number of bullet comments is used to indicate the number of bullet comments generated by users watching the live stream per unit of time. The schedule parameters and / or the event parameters are determined based on the average number of bullet comments.

4. The method according to claim 2, characterized in that, The step of determining at least one influencing parameter for the live broadcast of the i-th match corresponding to the i-th match in the n matches based on the bullet screen data includes: The average number of bullet comments is determined based on the bullet comment data. The average number of bullet comments is used to indicate the number of bullet comments generated by users watching the live stream per unit of time. Based on the average number of bullet comments, determine the team influence parameters for the live broadcast of the i-th match; Based on the team influence parameters and the match data of the i-th game, the live broadcast excitement level parameters are determined.

5. The method according to claim 1, characterized in that, The step of determining the live broadcast impact parameters associated with the i-th event live broadcast based on the at least one impact parameter includes: Based on the at least one influencing parameter, an average barrage prediction value is determined, which is used to indicate the predicted average barrage volume of the schedule to which the n events belong. The live broadcast impact parameters are determined based on the average predicted value of bullet comments and the average number of bullet comments in the i-th live broadcast.

6. The method according to any one of claims 1 to 5, characterized in that, The playback strategy for determining the future video playback includes at least one of the following: Determine the playback method for the future video; Determine the real-time monitoring method for the future video playback; The debugging method for the video playback device is determined, and the video playback device includes at least audio and video equipment and network support equipment.

7. A data processing apparatus, characterized in that, The device includes: The acquisition module is used to acquire the bullet screen data of n videos, which belong to the same series; the n videos are live videos of n events belonging to the same schedule, a schedule includes at least one event, one event corresponds to one live broadcast, and there are two teams competing in one event; The processing module is used to determine at least one influence parameter of the live broadcast of the i-th event corresponding to the i-th event in the n events based on the bullet screen data. The influence parameter is used to indicate the factor importance parameter corresponding to the quantitative indicator for playing the live broadcast of the i-th event. The processing module is further configured to determine the live broadcast impact parameter associated with the live broadcast of the i-th event based on the at least one impact parameter, wherein the live broadcast impact parameter is used to indicate the factor importance parameter corresponding to the non-quantitative indicator for playing the live broadcast of the i-th event. The processing module is further configured to predict the degree of abnormal impact of the non-quantitative indicators on future video playback based on the n live broadcast impact parameters corresponding to the n videos respectively, and determine the playback strategy for future video playback. Where n and i are positive integers greater than zero, and n is not less than i.

8. A computer device, characterized in that, The computer device includes a memory and a processor; The memory stores at least one piece of program code, which is loaded and executed by the processor to implement the data processing method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which is executed by a processor to implement the data processing method as described in any one of claims 1 to 6.

Images