Data processing method and apparatus, and nonvolatile storage medium

Abstract

The application discloses a data processing method and device and a nonvolatile storage medium. The method comprises the following steps: acquiring live time sequence data corresponding to a video live broadcast; classifying live time node data in which a preset condition occurs and live time node data in which the preset condition does not occur in a plurality of live time node data, and determining a live bandwidth demand index according to a classification result; determining a network quality characteristic quantization factor for representing network performance; determining a live value characteristic quantization factor for representing live value; and determining a comprehensive quantization evaluation index of live acceleration demand corresponding to the video live broadcast according to the live bandwidth demand index, the network quality characteristic quantization factor and the live value characteristic quantization factor. The application solves the technical problem that a live acceleration service cannot be accurately provided for a target user due to the inability to accurately determine the acceleration demand of different users for the network live broadcast.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data analysis, in particular to a data processing method and device and a nonvolatile storage medium. BACKGROUND

[0002] The rise and development of the live broadcast industry has experienced an industry budding stage starting from platform technology progress and traffic cost reduction, a wild development stage of live broadcast as a more direct information sharing way, a stage of rapid development of diversified forms of live broadcast popularized and integrated into the daily life of the general public, especially the integration with the e-commerce industry, making it a direct and effective medium for users in various industries to reach. At the same time, with the rapid development of the live broadcast industry and its promotion in different industries, the network speed of live broadcast accounts has become an important factor affecting the live broadcast effect. Therefore, it is particularly important to judge the network speed demand of live broadcast users and optimize the network speed in a timely manner according to the actual demand of live broadcast accounts.

[0003] Social development and technological progress can determine the demand for live broadcast through big data mining technology, but in combination with the actual situation, the existing technology still has many deficiencies.

[0004] Specifically, the traditional user demand identification method cannot accurately capture the real demand of users, especially in the live broadcast field, the demand of users changes at any time, and it is impossible to accurately identify target users who have high demand for network, smoothness, and stream stability, and thus it is impossible to accurately provide live broadcast acceleration services for target users.

[0005] In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0006] The present application provides a data processing method and device and a nonvolatile storage medium to at least solve the technical problem that it is impossible to accurately provide live broadcast acceleration services for target users due to the inability to accurately determine the acceleration demand of different users for network live broadcast.

[0007] According to an aspect of the present application, a data processing method is provided, comprising: obtaining live time series data corresponding to a video live broadcast, wherein the live time series data comprises a plurality of live time node data; classifying live time node data in which a preset condition occurs and live time node data in which the preset condition does not occur among the plurality of live time node data, and determining a live bandwidth demand index according to a classification result, wherein the preset condition comprises that a live quality index of the video live broadcast does not meet a preset requirement due to floating of a network bandwidth index; determining a network quality characteristic quantization factor for representing network performance according to a number of times of network problems occurring in a video live broadcast process; determining a live value characteristic quantization factor for representing live value according to a number of viewers of the video live broadcast and live revenue information; and determining a live acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast according to the live bandwidth demand index, the network quality characteristic quantization factor, and the live value characteristic quantization factor.

[0008] Optionally, classifying live time node data in which a preset condition occurs and live time node data in which the preset condition does not occur among the plurality of live time node data, and determining a live bandwidth demand index according to a classification result, comprises: taking a first preset time length as a time window, and counting live time node data in the time window; assigning a first numerical value to live time node data in which the preset condition occurs in the time window, and assigning a second numerical value to live time node data in which the preset condition does not occur in the time window, to obtain a first time series corresponding to the time window; obtaining live time node data in a second preset time length, wherein the second preset time length comprises a plurality of time windows, and performing merging processing on the first time series corresponding to the plurality of time windows in the second preset time length to obtain a target time series; calculating a mathematical expectation value of the target time series, and performing normalization processing on the mathematical expectation value to obtain a network quality level index in a video live broadcast process; taking the first preset time length as an independent variable, and the number of the first numerical value as a dependent variable, using a least square method to fit a linear equation to obtain a first target equation; mapping a first slope in the first target equation to a first digital interval by triangular transformation to obtain a network quality change trend index; and performing weighted summation on the network quality level index and the network quality change trend index to obtain the live bandwidth demand index.

[0009] Optionally, the network quality feature quantification factor used to represent the network performance is determined according to the number of network problems occurring in the video live broadcast process, including: obtaining the number of network problems occurring in the video live broadcast process in a first preset statistical period, wherein the first preset statistical period includes a plurality of first time steps; determining a first number of network problems occurring in each first time step; in order from the start time of the first preset statistical period to the end time, adding a weight coefficient to the first number corresponding to the i th first time step to obtain a plurality of second numbers, wherein the weight coefficient is the inverse of the Fibonacci number, the Fibonacci number F[i] corresponding to the i th first time step is F[i-1]+F[i-2], wherein F[0]=1, F[1]=1, and i is a positive integer greater than 1; summing the plurality of second numbers to obtain the network quality feature quantification factor.

[0010] Optionally, the live broadcast value feature quantification factor used to represent the live broadcast value is determined according to the number of viewers of the video live broadcast and the live broadcast revenue information, including: obtaining the number of viewers of the video live broadcast in a second preset statistical period, wherein the second preset statistical period includes a plurality of second time steps; determining a live broadcast viewer expectation value per unit second time step according to the ratio of the number of viewers to the number of second time steps; generating a second time sequence about the number of viewers corresponding to each second time step and the live broadcast revenue information in each second time step; using a least squares regression method to linearly regress and fit the second time sequence to obtain a second target equation; mapping a second slope in the second target equation to a second digital interval by a trigonometric transformation to obtain a revenue change trend indicator; and multiplying the live broadcast viewer expectation value per unit second time step and the revenue change trend indicator to obtain the live broadcast value feature quantification factor.

[0011] Optionally, the live broadcast acceleration demand comprehensive quantification evaluation index corresponding to the video live broadcast is determined, including: obtaining identification information of a target object performing the video live broadcast; determining whether the target object belongs to a multi-channel network organization according to the identification information; and multiplying the live broadcast acceleration demand comprehensive quantification evaluation index corresponding to the video live broadcast by a preset weight coefficient to obtain a target live broadcast acceleration demand comprehensive quantification evaluation index in the case where the target object belongs to the multi-channel network organization.

[0012] Optionally, the live broadcast acceleration demand comprehensive quantification evaluation index corresponding to the video live broadcast is determined according to the live broadcast bandwidth demand indicator, the network quality feature quantification factor, and the live broadcast value feature quantification factor, including: performing normalization processing on the live broadcast bandwidth demand indicator, the network quality feature quantification factor, and the live broadcast value feature quantification factor respectively, and performing multiplication calculation on the normalization processing results to obtain the live broadcast acceleration demand comprehensive quantification evaluation index corresponding to the video live broadcast.

[0013] Optionally, after determining the live streaming acceleration demand comprehensive quantitative evaluation index corresponding to the video live streaming, the method further comprises: dynamically adjusting network resource allocation of the live streaming service according to the live streaming acceleration demand comprehensive quantitative evaluation index.

[0014] According to still another aspect of the present application, a data processing apparatus is further provided, comprising: an acquisition module configured to acquire live streaming time sequence data corresponding to a video live streaming, wherein the live streaming time sequence data comprises a plurality of live streaming time node data; a first determination module configured to classify live streaming time node data in which a preset condition occurs and live streaming time node data in which the preset condition does not occur among the plurality of live streaming time node data, and determine a live streaming bandwidth demand index according to a classification result, wherein the preset condition comprises: a live streaming quality index of the video live streaming does not meet a preset requirement due to floating of a network bandwidth index; a second determination module configured to determine a network quality characteristic quantitative factor representing network performance according to a number of times of network problems occurring in a video live streaming process; a third determination module configured to determine a live streaming value characteristic quantitative factor representing live streaming value according to a number of viewers of the video live streaming and live streaming revenue information; and a fourth determination module configured to determine a live streaming acceleration demand comprehensive quantitative evaluation index corresponding to the video live streaming according to the live streaming bandwidth demand index, the network quality characteristic quantitative factor and the live streaming value characteristic quantitative factor.

[0015] According to still another aspect of the present application, a non-volatile storage medium is further provided, comprising a stored program, wherein the program controls a device in which the storage medium is located to execute the above data processing method when the program is run.

[0016] According to still another aspect of the present application, an electronic device is further provided, comprising: a memory and a processor, the processor being configured to run a program stored in the memory, wherein the program executes the above data processing method when the program is run.

[0017] According to still another aspect of the present application, a computer program is further provided, wherein the computer program is executed by a processor to implement the above data processing method.

[0018] According to still another aspect of the present application, a computer program product is further provided, comprising a non-volatile computer readable storage medium, wherein the non-volatile computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the above data processing method.

[0019] In the present application, the live time sequence data corresponding to the video live broadcast is acquired, wherein the live time sequence data includes a plurality of live time node data; the live time node data in which the preset condition occurs and the live time node data in which the preset condition does not occur are classified in the plurality of live time node data, and the live bandwidth demand index is determined according to the classification result, wherein the preset condition includes that the live quality index of the video live broadcast does not meet the preset requirement due to the floating of the network bandwidth index; the network quality characteristic quantization factor for representing the network performance is determined according to the number of network problems occurring in the video live broadcast process; the live value characteristic quantization factor for representing the live value is determined according to the number of viewers of the video live broadcast and the live income information; and the live acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast is determined according to the live bandwidth demand index, the network quality characteristic quantization factor and the live value characteristic quantization factor, so as to accurately determine the acceleration demand of different users for the network live broadcast, thereby realizing the technical effect of accurately providing the live acceleration service for the target user, and further solving the technical problem that the live acceleration service for the target user cannot be accurately provided due to the inability to accurately determine the acceleration demand of different users for the network live broadcast. BRIEF DESCRIPTION OF DRAWINGS

[0020] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute improper limitations on the present application. In the drawings:

[0021] Figure 1 is a flowchart of a data processing method according to an embodiment of the present application;

[0022] Figure 2 is a flowchart of another data processing method according to an embodiment of the present application;

[0023] Figure 3 is a structural diagram of a data processing device according to an embodiment of the present application;

[0024] Figure 4 is a hardware structural block diagram of a computer terminal of a data processing method according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0026] It should be noted that the terms "first", "second", and the like in the description and claims of the present application and above-described accompanying drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0027] In order to better understand the embodiments of the present application, the technical terms involved in the embodiments of the present application are explained as follows:

[0028] Forgetting curve: The forgetting curve is discovered by a German psychologist Ebbinghaus, which describes the law of human brain forgetting new things. Forgetting starts immediately after learning, and the process of forgetting is not uniform, the initial forgetting speed is very fast, and then gradually slows down.

[0029] Near effect: The near effect is a concept in psychology, which means that we are more likely to be influenced by recent experiences or information when making judgments or decisions, and ignore the long-term history or overall situation.

[0030] Multi-channel network (MCN): is any entity or organization that cooperates with content creators or directly produces various unique content, and performs business and operation functions on the network platform where the content is published. MCN is not affiliated with the platform owner or directly affiliated with the channel itself. MCN provides content planning and production, promotion, fan management, signing agency and other services for netizens and self-media.

[0031] In the related art, first, the user demand identification method cannot accurately capture the real needs of users, especially in the live broadcast field, the user's demand changes at any time, and it is impossible to accurately identify target users who have high demand for network, smoothness, and stream stability. Secondly, the user portrait construction method depends on limited data sources, resulting in an incomplete and inaccurate user portrait. Thirdly, the related operation strategy is relatively extensive, and cannot accurately operate and promote different user groups, resulting in unsatisfactory operation effect. In order to solve the above problems, the related solutions in the embodiments of the present application are provided, which are described in detail as follows.

[0032] According to the embodiment of the present application, a method embodiment of a data processing method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from here.

[0033] Figure 1 is a flowchart of a data processing method according to an embodiment of the present application, as shown in Figure 1 the method comprises the following steps:

[0034] Step S101, acquiring live time sequence data corresponding to the video live broadcast, wherein the live time sequence data includes a plurality of live time node data.

[0035] In step S101, by monitoring the video live broadcast stream in real time, the key performance indicators in the live broadcast process can be collected and recorded, including but not limited to live bandwidth, delay, packet loss rate, number of viewers, audience interaction frequency, live broadcast income, etc., forming a series of live time node data. These data are stored in the form of time sequence, which is convenient for subsequent analysis.

[0036] Step S102, classifying the live time node data in which the preset condition occurs and the live time node data in which the preset condition does not occur among the plurality of live time node data, and determining the live bandwidth demand index according to the classification result, wherein the preset condition includes that the live quality index of the video live broadcast does not meet the preset requirement due to the floating of the network bandwidth index.

[0037] Among them, the network bandwidth index includes but is not limited to the following dimensions: upload bandwidth, download bandwidth, packet loss rate, delay, jitter. The network bandwidth index in step S102 can be any one of the above dimensions, or a weighted combination of multiple dimensions. The live quality index includes but is not limited to: video resolution, video frame rate, interaction delay. The live quality index in step S102 can be any one of the above dimensions, or a weighted combination of multiple dimensions.

[0038] For example, for the preset minimum video resolution, in high-definition live streaming, the preset minimum resolution can be set to 720p. When the network bandwidth decreases to insufficient to maintain this resolution, the live streaming quality is considered to be substandard. For the preset maximum delay time, the live streaming delay should be kept within 2 seconds to ensure the audience's sense of real-time. Once the delay exceeds the preset value, the live streaming quality can be considered to be affected. For the preset video buffering frequency and duration: for example, buffering no more than once every 5 minutes and no more than 5 seconds each time is acceptable. Buffering events exceeding this frequency and duration are considered to be a decrease in live streaming quality. For the preset audio synchronization error, a synchronization error of less than 100 milliseconds is acceptable, and exceeding this value can affect the live streaming experience. For the preset packet loss rate: a packet loss rate of no more than 1% is a basic requirement for live streaming services. An increase in packet loss rate means a decrease in transmission efficiency and video quality.

[0039] Step S103, according to the number of network problems occurring in the video live streaming process, determine the network quality characteristic quantization factor for characterizing network performance.

[0040] Among them, the delay time: usually in milliseconds (ms), represents the average time from sending to receiving data. The preset requirement can be less than 100 ms, otherwise it is considered as network problem. Packet loss rate: represents the proportion of data packets that fail to arrive successfully during network transmission, usually expressed in percentage. For example, a packet loss rate of more than 1% can be considered as network problem. Maximum bandwidth and average bandwidth: represent the maximum and average data transmission rate that the network can provide, respectively, in Mbps (megabits per second). When the actual bandwidth of the network is lower than the required minimum bandwidth threshold, it can be considered as network problem. Jitter: time fluctuation of network delay, usually also in milliseconds. For example, jitter exceeding 20 ms can be considered as a problem. Encryption error rate: the proportion of data packets that are not correctly encrypted or decrypted during data transmission, the preset encryption success rate is 100%, any value lower than this can be considered as network problem. Connection failure rate: the proportion of attempts to establish network connection but failed, the preset connection success rate is 100%, a connection failure rate exceeding a certain threshold (such as 0.5%) can indicate network instability.

[0041] Step S104, according to the number of viewers of the video live streaming and the live streaming revenue information, determine the live streaming value characteristic quantization factor for characterizing live streaming value.

[0042] Step S105, according to the live streaming bandwidth demand index, the network quality characteristic quantization factor and the live streaming value characteristic quantization factor, determine the live streaming acceleration demand comprehensive quantization evaluation index corresponding to the video live streaming.

[0043] According to the above steps, the live broadcast time sequence data corresponding to the video live broadcast is acquired, wherein the live broadcast time sequence data includes a plurality of live broadcast time node data; the live broadcast time node data in which the preset condition occurs and the live broadcast time node data in which the preset condition does not occur are classified in the plurality of live broadcast time node data, and a live broadcast bandwidth demand index is determined according to the classification result, wherein the preset condition includes that the live broadcast quality index of the video live broadcast does not meet the preset requirement due to the floating of the network bandwidth index; a network quality characteristic quantization factor for representing the network performance is determined according to the number of network problems occurring in the video live broadcast process; a live broadcast value characteristic quantization factor for representing the live broadcast value is determined according to the number of viewers of the video live broadcast and live broadcast income information; and a live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast is determined according to the live broadcast bandwidth demand index, the network quality characteristic quantization factor and the live broadcast value characteristic quantization factor, so as to accurately determine the acceleration demand of different users for the network live broadcast, thereby achieving the technical effect of accurately providing live broadcast acceleration services for target users.

[0044] The steps shown in the following are exemplarily described and explained. Figure 1

[0045] According to some optional embodiments of the present application, the live broadcast time node data in which the preset condition occurs and the live broadcast time node data in which the preset condition does not occur are classified in the plurality of live broadcast time node data, and a live broadcast bandwidth demand index is determined according to the classification result, which can be realized by the following method: taking a first preset time length as a time window, the live broadcast time node data in the time window is counted; a first value is given to the live broadcast time node data in which the preset condition occurs in the time window, and a second value is given to the live broadcast time node data in which the preset condition does not occur in the time window, to obtain a first time sequence corresponding to the time window; the live broadcast time node data in a second preset time length is acquired, wherein the second preset time length includes a plurality of time windows, the first time sequences corresponding to the plurality of time windows in the second preset time length are processed by merging to obtain a target time sequence; the mathematical expectation value of the target time sequence is calculated, and the mathematical expectation value is normalized to obtain a network quality level index in the video live broadcast process; taking the first preset time length as the independent variable and the number of the first value as the dependent variable, a linear equation is fitted using the least square method to obtain a first target equation; the first slope in the first target equation is mapped into a first digital interval by triangular transformation to obtain a network quality change trend index; the network quality level index and the network quality change trend index are weighted and summed to obtain the live broadcast bandwidth demand index.

[0046] For example, D represents whether there is a demand for live broadcast acceleration, and 24 hours is taken as a time window to count the live broadcast time node data in the time window.

[0047] ​In the time window, if the live time node data appears abnormal due to bandwidth, it means that the current bandwidth cannot meet the live demand, at this time D is assigned a value of 1, otherwise D is assigned a value of 0. Based on this, add the time variable t, then construct the time series {D t} that conforms to the "0-1" distribution. Because the bandwidth demand usually appears as an abnormal phenomenon, most of the records of {D t} are 0. By counting multiple 24-hour time series {D t}, the target time series {D d} is obtained. By calculating the expectation E value of the target time series {D d}, the size of the live bandwidth demand can be quantified, and the E value is positively correlated with the bandwidth demand. If the E value is 0, it means that the bandwidth has no effect on the live. The calculation formula of the expectation E is as follows:

[0048]

[0049] The value range of the abnormal expectation E calculated according to formula (1) in the statistical time is [0, +∞), considering that if the live abnormality occurs more than a certain number of times in a day, it will have a disastrous impact on the live, therefore, the data with E value greater than 100 is uniformly assigned a value of 100, the value range of the abnormal number expectation E is scaled to the interval [0, 100), and in order to facilitate subsequent integration and calculation, the value range of E is scaled to the interval [0, 1) through logarithmic transformation, denoted as EG.

[0050] The calculation formula is as follows:

[0051] EG=log 100 E (2)

[0052] For the target time series {D d}, take time d as the independent variable and abnormal number D d as the dependent variable, n1 as the sequence length, use the least squares method to fit a linear equation, denote the slope as b, and by analyzing the slope b of the linear equation, the trend of the time series {D d} can be determined. If the slope b is positive, it means that the time series is in an upward trend. If the slope b is negative, it means that the time series is in a downward trend. When the slope b is close to zero, it means that the time series may have no obvious trend, and the change amplitude can also be determined according to the absolute value of b. The specific calculation formula of the slope b is as follows:

[0053]

[0054] Because the value range of the slope b is (-∞, +∞), in order to eliminate the dimensional influence between different indicators during fusion, make the data comparable, and facilitate subsequent analysis, the value range of b is mapped to the interval (-1, 1) through a triangular transformation, denoted as LB, and the calculation formula is as follows:

[0055]

[0056] Because both the network quality level EG and the network quality trend LB will very obviously affect the anchor's perception of network quality and trigger the anchor's demand for live streaming acceleration, the weighted average form is used for calculation to obtain the live streaming bandwidth demand EGL, and the formula is as follows:

[0057]

[0058] According to some optional embodiments of the present application, the network quality characteristic quantization factor for representing network performance is determined according to the number of network problems occurring in the video live streaming process, which can be achieved by the following method: obtaining the number of network problems occurring in the video live streaming process in a first preset statistical period, wherein the first preset statistical period includes a plurality of first time steps; determining a first number of network problems occurring in each first time step; adding a weight coefficient to the first number corresponding to the i-th first time step in order from the start time to the end time of the first preset statistical period, to obtain a plurality of second numbers, wherein the weight coefficient is the reciprocal of the Fibonacci number, the Fibonacci number F[i] corresponding to the i-th first time step is F[i-1]+F[i-2], wherein F[0]=1, F[1]=1, and i is a positive integer greater than 1; summing the plurality of second numbers to obtain the network quality characteristic quantization factor.

[0059] In the above embodiments, first, the preset statistical period and the time step are determined. The first preset statistical period is set, for example, the first 10 minutes of the entire live streaming process is taken as the statistical period. The first time step is divided into a plurality of first time steps, each of which is 1 minute, in order to more carefully monitor the network condition.

[0060] Secondly, the network problem can include uploading bandwidth drop, abnormal packet loss rate, significant increase in delay, etc. In each first time step (for example, each minute in the last 10 minutes), it is detected whether the above defined network problem occurs in the video live streaming process. If it occurs, it is recorded as the first time step in which the network problem occurs; otherwise, it is recorded as the first time step in which the network problem does not occur. Further, the number of network problems occurring in each first time step, i.e., the first number, is counted.

[0061] Again, each time step within the network problem occurs (i.e. the first number) is assigned a weight coefficient, the weight coefficient is the inverse of F[i], that is, 1 / F[i]. According to the first preset statistical period from the beginning to the end of the order, from the first minute to the end of the tenth minute, each first time step within the network problem occurs is weighted one by one.

[0062] Assuming the first preset statistical period is the first 10 minutes after the video live starts, each first time step is 1 minute. The first 10 values of the Fibonacci sequence are used as the denominator of the weight coefficient, and the first 10 values of the Fibonacci sequence are defined as follows: F[0]=1, F[1]=1, F[2]=2, F[3]=3, F[4]=5, F[5]=8, F[6]=13, F[7]=21, F[8]=34, F[9]=55.

[0063] In each first time step, record the number of network problems that occur during the video live (the first number). Assuming the number of network problems in the statistical period is as follows: the first minute (F[1]=1): 2 network problems occur; the second minute (F[2]=2): 1 network problem occurs; the third minute (F[3]=3): 3 network problems occur; the fourth minute (F[4]=5): 2 network problems occur; the fifth minute (F[5]=8): 4 network problems occur; the sixth minute (F[6]=13): 1 network problem occurs; the seventh minute (F[7]=21): 2 network problems occur; the eighth minute (F[8]=34): 3 network problems occur; the ninth minute (F[9]=55): 1 network problem occurs.

[0064] Then for the first minute (F[1]=1): the second number=2 / 1=2; for the second minute (F[2]=2): the second number=1 / 2=0.5; for the third minute (F[3]=3): the second number=3 / 3=1; for the fourth minute (F[4]=5): the second number=2 / 5=0.4; for the fifth minute (F[5]=8): the second number=4 / 8=0.5; for the sixth minute (F[6]=13): the second number=1 / 13≈0.077; for the seventh minute (F[7]=21): the second number=2 / 21≈0.095; for the eighth minute (F[8]=34): the second number=3 / 34≈0.088; for the ninth minute (F[9]=55): the second number=1 / 55≈0.018.

[0065] Finally, the sum of the second number is calculated, that is, the network quality characteristic quantization factor can be obtained.

[0066] It should be noted that, unlike the live attribute related problems, the live host can intuitively feel the negative impact of the related problems on the ongoing live broadcast. The network factor mainly shows that the network backend data in the live process cannot provide the network resources required for live broadcast, such as network congestion, high packet loss rate, network delay, network jitter, and other network quality problems. The network quality problem may not affect the current live effect, but long-term mismatch of resources will eventually lead to a decrease in the live satisfaction of the live host. Therefore, network related problems are one of the main reasons for the adverse effects of live broadcast, that is, the root cause of affecting the live picture sense in addition to the individual differences of the live host and the hardware equipment he equipped. Therefore, by quantifying the network quality characteristics, the live broadcast demand can be more accurately predicted. Because occasional network delay and other quality problems are not obvious to the live host, only when the related problems occur relatively frequently will they affect the live host. Therefore, the frequency of network quality problems, that is, the number of network problems, can be used to quantify the network quality factor. At the same time, considering that the individual live host makes related decisions mainly based on his own judgment, according to the theory of Ebbinghaus forgetting curve and the theory of "near effect" in psychology, the live host individual memory follows the forgetting curve, that is, the longer the distance from the current time, the weaker the related memory, and the behavior has "near effect", that is, when making decisions, it is more easily influenced by recent events or information. Therefore, when calculating the frequency, the influence factors of the forgetting curve time span and "near effect" on the live host's decision need to be corrected.

[0067] Because different network problems will affect the live host, in the counting process, no matter what problem occurs, it is included in the statistics, and the number of network problems occurring in a fixed time step is N i Where N is the number of problems that occur, i is a certain time step in the statistical period, in order to consider the "forgetting curve" and "near effect", according to the network problem occurrence distance from the statistical time from near to far, add the reciprocal of the Fibonacci number F(i) (F[i] = F[i-1] + F[i-2] (i >= 2, F[0] = 1, F[1] = 1)) as the weighting coefficient, denoted as That is, add the coefficient to construct the network quality characteristic quantization factor, then perform weighted sum, and denote the network quality characteristic quantization factor as NF. The calculation formula is as follows:

[0068]

[0069] In some optional embodiments of the present application, the live streaming value characteristic quantification factor used to represent the live streaming value is determined according to the number of viewers of the video live streaming and the live streaming revenue information, which can be achieved by the following method: obtaining the number of viewers of the video live streaming in a second preset statistical period, wherein the second preset statistical period includes a plurality of second time steps; determining the expected value of the number of live streaming viewers per unit second time step according to the ratio of the number of viewers to the number of second time steps; generating a second time sequence about the number of viewers corresponding to each second time step and the live streaming revenue information in each second time step; performing linear regression fitting on the second time sequence using the least squares regression method to obtain a second target equation; mapping the second slope in the second target equation to a second digital interval by triangular transformation to obtain a revenue change trend index; and performing multiplication calculation on the expected value of the number of live streaming viewers per unit second time step and the revenue change trend index to obtain the live streaming value characteristic quantification factor.

[0070] It can be understood that the live streaming value LSR (Live streaming revenue) is the actual benefit value brought by the live streaming account to the live streaming host, for example, the number of account fans, the number of likes, the number of plays and other live streaming data that are positively correlated with the platform commission and the live streaming sales, which can indirectly reflect the value of the live streaming account and embody the value of the account produced by the live streaming host. Therefore, the size of the related value will directly affect the willingness of the live streaming host to invest costs. For example, if the stable revenue value of the live streaming account is very large or the revenue value has been increasing positively, and the expected future revenue value is very considerable, then the live streaming host is not sensitive to the network resource cost investment for improving the quality of the live streaming account and increasing audience satisfaction; on the contrary, the willingness of the live streaming host to improve the performance of the account is not high. Therefore, the cost investment sensitivity of the live streaming host can be quantified by constructing a live streaming revenue characteristic quantification factor.

[0071] According to the above analysis, two points need to be considered for quantifying and analyzing the live streaming revenue. One is the stable revenue value of the live streaming account of the live streaming host, which is the main influencing factor of the cost investment of the live streaming host; the other is the growth of the live streaming account, i.e., the quantification of the change trend of the value of the live streaming account, which can also judge the willingness of the live streaming host to invest costs. For the quantification of the stable revenue value of the live streaming account of the live streaming host, the number of viewers of the live streaming account in the last time can be used for evaluation. However, the number of viewers of a single live streaming is easily affected by abnormal events and fluctuates greatly, which cannot be used to evaluate the influence of the stable value of the live streaming account. Therefore, in order to eliminate the influence of abnormal values on the evaluation results, the expected value of the number of viewers of the live streaming account in the last six months is calculated to quantitatively evaluate the revenue stability of the live streaming account, denoted as value expectation IE. The statistical period includes M months, the statistical month is m, and the number of viewers corresponding to the statistical month is I mThe expected calculation formula of the live broadcast account monthly live broadcast watching number in the past six months is:

[0072]

[0073] The growth of the live broadcast account of the live broadcast host, i.e., the change trend of the watching number and the like, is also an important influencing factor of the cost input willingness of the live broadcast host, but the relevant change trend can be used as an account value expectation factor rather than an explicit factor, and in general, the account value cannot be evaluated alone. However, the watching number index change trend as an expectation factor can be used as a weighted item to modify the quantitative evaluation of the live broadcast account value, and improve the accuracy of the quantitative evaluation of the live broadcast value characteristic factor. Next, the value trend of the live broadcast account is quantitatively calculated.

[0074] In order to quantitatively calculate the value trend of the live broadcast account over time ICT (Income change trend), a live broadcast account monthly watching number change time sequence {(x, y)} is constructed, where x is the statistical month, y is the live broadcast income in the statistical month, n3 is the total number of months included in the statistical period, a linear regression fitting is performed on the data set using the least square method, the slope of the fitting equation is calculated as ICT, and then according to the "positive and negative" of ICT, it is judged whether the live broadcast income changes over time is increasing or decreasing trend, and according to the absolute value of ICT, the change amplitude of the live broadcast watching number can be judged, and the calculation formula is as follows:

[0075]

[0076] Because the value range of the slope Ib is (-∞, +∞), it cannot be directly used as a coefficient to modify the value of the live broadcast account, therefore, through a triangular transformation, the value range of b is mapped to (-1, 1), and the normalized calculation value is used as the coefficient of the influence factor of the live broadcast account value change, and the coefficient calculation formula is as follows:

[0077]

[0078] Considering that if the value of Ib1 is zero under the condition that the live broadcast account value is a stationary time sequence, the quantitative evaluation value of the stability of the live broadcast account value will be 0. Therefore, by adding 1 to Ib, the value range is mapped to the range of (0, 2), and then multiplied by the income expectation IE to obtain the final live broadcast value characteristic quantitative factor, and the calculation formula is as follows:

[0079] LSR = (1 + Ib) IE (10)

[0080] As some optional embodiments of the present application, the method for determining the live streaming acceleration demand comprehensive quantitative evaluation index corresponding to the video live streaming can be implemented as follows: obtaining identification information of a target object performing video live streaming; determining whether the target object belongs to a multi-channel network organization according to the identification information; and multiplying the live streaming acceleration demand comprehensive quantitative evaluation index corresponding to the video live streaming by a preset weight coefficient to obtain a target live streaming acceleration demand comprehensive quantitative evaluation index in the case where the target object belongs to the multi-channel network organization.

[0081] In addition, the method for determining the live streaming acceleration demand comprehensive quantitative evaluation index corresponding to the video live streaming according to the live streaming bandwidth demand index, the network quality characteristic quantitative factor and the live streaming value characteristic quantitative factor can be implemented as follows: performing normalization processing on the live streaming bandwidth demand index, the network quality characteristic quantitative factor and the live streaming value characteristic quantitative factor respectively, and performing multiplication calculation on the normalization processing results to obtain the live streaming acceleration demand comprehensive quantitative evaluation index corresponding to the video live streaming.

[0082] It should be noted that through analysis of the live streaming industry scene, it is found that there are mainly two kinds of live streaming ecologies, one of which is an independent live streamer who does not join a MCN organization. Since all live streaming process related activities are independently responsible, the main reason affecting the live streaming acceleration demand degree is that the live streamer encounters a lag during live streaming, which affects the interaction with the live streaming audience, and then affects the viewing experience and satisfaction of the live streaming audience, and finally affects the income obtained by the live streamer during the live streaming process. From a technical point of view, the main reason for the appearance of lag in live streaming is that the short-term uplink and downlink data traffic is too large, while the original network basic resources cannot meet the demand, causing network congestion.

[0083] In order to more comprehensively obtain the factors affecting the live streaming acceleration demand of the live streamer, the specific scene is decomposed and refined, and the factor characteristics are converged from three aspects of live streaming attributes, network quality and live streaming income. The live streaming attributes are composed of live streaming software DPI data and platform side user data, including live streaming platform fan quantity, live streaming frequency, audience interaction times, audience quantity and other characteristics. This part of characteristics mainly shows the live streaming bandwidth demand degree, for example, the audience quantity is too large during live streaming, which causes mismatch with the network bandwidth, and may cause network lag and congestion, which needs to be solved by increasing the bandwidth. The buffer duration, buffer times and picture definition in the network quality characteristics show the actual network quality problems in live streaming, which need to be improved in time, otherwise the live streaming viewing comfort and smoothness will be affected, and then the audience satisfaction will be affected. The fan quantity, like quantity and play quantity of the anchor live streaming account reflect the live streaming account popularity related characteristics, which can reflect the value of the live streaming account.

[0084] Another live broadcast ecology is dependent on live broadcast base, net red incubation base, and specific live broadcast industry group acceleration demand. In this live broadcast ecology, in addition to considering the demand of individual live broadcast host for acceleration, the entire live broadcast group industry attribute, the region to which it belongs, and the stage live broadcast acceleration demand are also considered.

[0085] According to the quantitative indicators calculated for different live broadcast acceleration demand scenarios according to the foregoing steps, the demand degree of individual live broadcast host and live broadcast group for live broadcast acceleration service is calculated.

[0086] 1. Acceleration demand quantitative evaluation and analysis model in individual live broadcast host live broadcast scenario.

[0087] The individual live broadcast host live broadcast scenario constructs three influencing factors, live broadcast bandwidth demand degree quantitative indicator EGL, network quality characteristic quantitative factor NF, and live broadcast income characteristic quantitative factor LSR. Because the dimensions of the three influencing factors are different, they cannot be compared and combined. In order to eliminate the dimensional influence, the Z-score normalization method is used to map the value range of the three influencing factors to the (-1, 1) space, and the normalized influencing factor indicators EGL nor , NF nor , and LSR nor are obtained. The sum of the three influencing factors is calculated to obtain the average value, and the live broadcast acceleration demand comprehensive quantitative evaluation index Single_LSA in the individual live broadcast host live broadcast scenario is obtained. The calculation formula is as follows:

[0088]

[0089] 2. Acceleration demand quantitative evaluation and analysis model in live broadcast host scenario dependent on live broadcast base, net red incubation base, and specific live broadcast industry.

[0090] Compared with individual live broadcasters, live broadcasters attached to live bases, Internet celebrity incubation bases and specific live industries generally depend on the decisions of live base managers for network acceleration services. The live base managers not only consider the needs of individual live broadcasters for acceleration, but also comprehensively consider the industry attributes of the entire live group, the regions to which the live group belongs and the live acceleration needs of the live group for stage plans. Through analysis of these factors, it is found that the main influencing factor is the industry attribute of the live group, which is divided into two cases. One is a public welfare institution, which is generally not income value-oriented and is relatively insensitive to price factors. When purchasing live acceleration services, it is more concerned about whether the desired live promotion effect can be achieved. The other is a commercial live industry, which is mainly income value-oriented and considers the overall income of the live group. If the overall income does not meet the income expectation, the possibility of purchasing live acceleration services is small, and vice versa. According to this situation, the possibility of purchasing live acceleration services by public welfare institutions is greater, while the possibility of purchasing live acceleration services by commercial enterprises is small. Therefore, a weight coefficient β is added for correction. If it is a public welfare institution, the coefficient is assigned a value of 1, and if it is a commercial enterprise, the coefficient is assigned a value of 0.8.

[0091] For the live acceleration needs of the live broadcaster group attached to the group Group_LSA, an industry coefficient β is also needed to be added for weighting and quantification. The calculation formula is as follows:

[0092] Group_LSA = β * Single_LSA (12)

[0093] According to the size of single_LSA and Group_LSA, the live acceleration needs of the live broadcaster in the live scene can be determined. The larger the value, the higher the demand, and vice versa. In actual operation, customers can be operated and recommended in batches and multiple gradients according to the size of single_LSA and Group_LSA to improve the operation success rate and reduce the operation cost.

[0094] As some optional embodiments of the present application, after determining the live acceleration demand comprehensive quantitative evaluation index corresponding to the video live, the following steps can also be performed: dynamically adjusting the network resource allocation of the live service according to the live acceleration demand comprehensive quantitative evaluation index.

[0095] Suppose the current live streaming platform monitors the LSAD-CQI of two live streaming accounts A and B as 8.5 and 4.2 respectively. Account A has higher live streaming bandwidth requirements, frequent network problems, and high viewership and revenue, while account B has relatively weak characteristics in these aspects. Based on this, the live streaming platform decides to provide more network resources for account A, including increasing uplink bandwidth, optimizing CDN node distribution, and increasing server cache, to ensure the smoothness of account A's live streaming and user satisfaction. At the same time, the platform keeps the network resource allocation for account B unchanged, but continues to monitor its LSAD-CQI for subsequent adjustments.

[0096] From the above steps, the present application has the following advantages: 1. Rich and high-quality data sources: comprehensive consideration of operator backend network data, live streaming platform network quality data, and live streaming platform user behavior data. In particular, the operator backend network data contains quasi-real-time data of network quality changes. Compared with traditional live streaming platform data or network backend data, it has better data support in live streaming acceleration demand evaluation, making the analyzed live streaming acceleration demand more accurate. 2. More comprehensive acceleration demand scenario analysis: In the evaluation process of live streaming acceleration demand scenarios, the live streaming acceleration demand of multiple scenarios is quantitatively evaluated from the two scenarios of single live streaming host and institutionally managed live streaming host, realizing the differentiated evaluation and analysis of live streaming acceleration of different live streaming host groups, and providing a theoretical basis for the formulation of differentiated operation policies.

[0097] Figure 2 is a flowchart of another data processing method according to an embodiment of the present application, as shown in Figure 2 , the method comprises the following steps:

[0098] Step S201: demand analysis. First, identify the acceleration demand of live streaming in different scenarios. Key factors include: 1. Live streaming attributes: including live streaming type (such as game, education, entertainment, etc.), live streaming duration, live streaming time period (peak or non-peak period), etc. 2. Network quality: refers to network problems encountered during live streaming, such as packet loss rate, delay, bandwidth fluctuation, etc. 3. Live streaming revenue situation: involves the revenue of live streaming host, viewership, user interaction (such as comments, likes), etc., reflecting the commercial value and user engagement of live streaming.

[0099] Step S202: construct individual live streaming factor feature quantization factor. Based on the results of demand analysis, construct individual live streaming factor feature quantization factor, which comprehensively considers the specific attributes of live streaming, such as live streaming type, time period, etc., to provide a preliminary quantization index for each live streaming scenario for subsequent analysis.

[0100] Step S203: Construct network quality characteristic quantification factor. Network quality characteristic quantification factor is constructed according to the frequency of network problems in live broadcast process. Specifically, it includes: network problem monitoring: real-time detection of network quality in live broadcast process, and recording the number of problems in a specific time window (such as every minute).

[0101] Step S204: Construct live broadcast revenue characteristic quantification factor.

[0102] Live broadcast revenue characteristic quantification factor involves analysis of the commercial value and user engagement of live broadcast. The specific steps include: 1. Consider the fan base, number of viewers, conversion rate, etc. of the live broadcast host, and evaluate the current value of the live broadcast account. 2. Track the revenue trend and user growth trend of the live broadcast account, and evaluate the long-term growth potential of the account. 3. Quantify the value and growth of the live broadcast account, and establish a characteristic quantification factor reflecting the revenue status of live broadcast, which is used to evaluate the potential commercial return of live broadcast acceleration service.

[0103] Step S205: Construct live broadcast acceleration demand quantification evaluation analysis model in multiple scenarios.

[0104] In the model construction stage, the above three quantification factors (individual live broadcast factor characteristics, network quality characteristics, and live broadcast revenue characteristics) are weighted and summed, and the mean value is calculated to obtain the comprehensive quantification evaluation index of live broadcast acceleration demand.

[0105] Obtain the identification information of the target object of the ongoing video live broadcast from the background database of the live broadcast platform. This may include the unique ID of the live broadcast account, the ID of the affiliated organization, etc. to facilitate subsequent attribution judgment and demand evaluation. According to the identification information, query the registration information or contract terms of the target object to determine whether the live broadcast account belongs to an MCN agency. MCN agencies usually have special cooperation with live broadcast platforms, and the live broadcast accounts under their agency may enjoy priority in resources or preferential policies. For target objects determined to belong to MCN agencies, we will adjust the live broadcast acceleration demand comprehensive quantification evaluation index (LSAD-CQI) and the preset weight coefficient to obtain the target live broadcast acceleration demand comprehensive quantification evaluation index.

[0106] Suppose there are three live broadcast scenarios: Scenario A: an educational live broadcast with good live broadcast attributes, average network quality, and excellent revenue situation. Scenario B: a game live broadcast with excellent live broadcast attributes, poor network quality, and good revenue situation. Scenario C: an entertainment live broadcast with average live broadcast attributes, good network quality, and average revenue situation.

[0107] Through analysis and quantification of the three characteristic factors in each scenario, the following values are obtained: Scene A: (AIF = 80), (NQF = 50), (LRF = 90). Scene B: (AIF = 90), (NQF = 30), (LRF = 70). Scene C: (AIF = 50), (NQF = 80), (LRF = 50).

[0108] According to the model output calculation formula, the following comprehensive quantitative evaluation indicators are calculated: LSAD-CQI A = 73.33; LSAD-CQI B = 63.33; LSAD-CQI C = 60.

[0109] Based on multi-scenario analysis, by constructing the LSAD-CQI model, the three key factors of live streaming properties, network quality, and live streaming revenue situation can be comprehensively considered, providing precise quantitative evaluation indicators for live streaming acceleration services. The platform can dynamically adjust network resource allocation according to the level of LSAD-CQI value, and preferentially meet the live streaming scenarios with higher LSAD-CQI value, thereby optimizing the overall live streaming experience and resource utilization efficiency. This method not only improves the pertinence of live streaming acceleration services, but also promotes the healthy development of the live streaming industry, meeting the needs of users for high-quality live streaming experience.

[0110] From the above steps, the application fully analyzes the actual live streaming scenarios, constructs corresponding live streaming acceleration scenario demand quantitative evaluation models from the perspectives of individual live streamers and organized live streamers, and realizes differentiated evaluation and quantification of live streaming acceleration demand in different scenarios. Through the live streaming acceleration demand quantitative evaluation model, the previous qualitative analysis of live streaming acceleration demand relying mainly on subjective experience and simple statistical analysis is changed, the results are more reliable and scientific, and the accuracy of live streaming acceleration service evaluation results is effectively improved. In addition, after scoring by the quantitative model, the result is more convenient and effective in result landing. The live streaming acceleration demand quantitative evaluation model scores and evaluates the demand degree of each live streamer, and the higher the score value, the higher the demand degree, and vice versa, realizing the display of the live streaming acceleration demand degree of the live streamer in a more intuitive way. The operation personnel can formulate multi-gradient operation strategies for different influence degrees according to the size of the live streaming acceleration demand score of the live streamer, construct differentiated operation capabilities, and realize precise allocation of operation resources.

[0111] Figure 3 is a structural diagram of a data processing device according to an embodiment of the application, as shown in Figure 3 The device comprises:

[0112] The acquisition module 31 is configured to acquire live time sequence data corresponding to the video live broadcast, wherein the live time sequence data comprises a plurality of live time node data.

[0113] The first determination module 32 is configured to classify, in the plurality of live time node data, live time node data in which the preset condition occurs and live time node data in which the preset condition does not occur, and determine a live bandwidth demand index according to a classification result, wherein the preset condition comprises that a live quality index of the video live broadcast does not meet a preset requirement due to floating of a network bandwidth index.

[0114] The second determination module 33 is configured to determine a network quality characteristic quantization factor for representing network performance according to a number of times of network problems occurring in the video live broadcast process.

[0115] The third determination module 34 is configured to determine a live value characteristic quantization factor for representing live value according to a number of viewers of the video live broadcast and live revenue information.

[0116] The fourth determination module 35 is configured to determine a live acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast according to the live bandwidth demand index, the network quality characteristic quantization factor, and the live value characteristic quantization factor.

[0117] Optionally, the first determination module 32 is further configured to perform the following steps: taking a first preset time length as a time window, counting live time node data in the time window; assigning a first numerical value to live time node data in which the preset condition occurs in the time window, and assigning a second numerical value to live time node data in which the preset condition does not occur in the time window, to obtain a first time sequence corresponding to the time window; acquiring live time node data in a second preset time length, wherein the second preset time length comprises a plurality of time windows, performing merging processing on the first time sequences corresponding to the plurality of time windows in the second preset time length to obtain a target time sequence; calculating a mathematical expectation value of the target time sequence, and performing normalization processing on the mathematical expectation value to obtain a network quality level index in the video live broadcast process; using a least square method to fit a linear equation with the first preset time length as an independent variable and a number of the first numerical value as a dependent variable, to obtain a first target equation; mapping a first slope in the first target equation to a first digital interval by triangular transformation to obtain a network quality change trend index; and performing weighted summation on the network quality level index and the network quality change trend index to obtain the live bandwidth demand index.

[0118] Optionally, the second determining module 33 is further configured to perform the following steps: obtaining the number of network problems in the video live broadcast process in a first preset statistical period, wherein the first preset statistical period comprises a plurality of first time steps; determining a first number of network problems in each first time step; adding a weight coefficient to the first number corresponding to the i th first time step in order from the start time of the first preset statistical period to the end time to obtain a plurality of second numbers, wherein the weight coefficient is the reciprocal of the Fibonacci number, the Fibonacci number F[i] corresponding to the i th first time step is F[i-1]+F[i-2], wherein F[0]=1, F[1]=1, and i is a positive integer greater than 1; and summing the plurality of second numbers to obtain a network quality characteristic quantization factor.

[0119] Optionally, the third determining module 34 is further configured to perform the following steps: obtaining the number of viewers of the video live broadcast in a second preset statistical period, wherein the second preset statistical period comprises a plurality of second time steps; determining a live broadcast viewer expectation value per unit second time step according to the ratio of the number of viewers to the number of second time steps; generating a second time sequence of the number of viewers corresponding to each second time step and live broadcast revenue information in each second time step; performing linear regression fitting on the second time sequence using a least squares regression method to obtain a second target equation; mapping a second slope in the second target equation to a second digital interval by triangular transformation to obtain a revenue change trend indicator; and performing multiplication calculation on the live broadcast viewer expectation value per unit second time step and the revenue change trend indicator to obtain a live broadcast value characteristic quantization factor.

[0120] Optionally, the fourth determining module 35 is further configured to perform the following steps: obtaining identification information of a target object performing video live broadcast; determining whether the target object belongs to a multi-channel network organization according to the identification information; and multiplying the live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast by a preset weight coefficient to obtain a target live broadcast acceleration demand comprehensive quantization evaluation index in the case where the target object belongs to the multi-channel network organization.

[0121] Optionally, the fourth determining module 35 is further configured to perform the following steps: performing normalization processing on the live broadcast bandwidth demand indicator, the network quality characteristic quantization factor, and the live broadcast value characteristic quantization factor respectively, and performing multiplication calculation on the normalization processing results to obtain the live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast.

[0122] Optionally, the data processing apparatus is further configured to perform the following steps after determining the live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast: dynamically adjusting network resource allocation of the live broadcast service according to the live broadcast acceleration demand comprehensive quantization evaluation index.

[0123] It should be noted that each of the modules in the above Figure 3 may be a program module (for example, a program instruction set for implementing a certain specific function) or a hardware module, and for the latter, it can be in the form of, but not limited to: each of the above modules is in the form of a processor, or the functions of the above modules are implemented by a processor.

[0124] It should be noted that the preferred embodiments of the embodiments shown in Figure 3 will not be described here. Figure 1

[0125] Figure 4 A hardware structure block diagram of a computer terminal for implementing a data processing method is shown. As shown in Figure 4 , the computer terminal 40 can include one or more (shown in the figure as 402a, 402b, …, 402n) processors 402 (the processor 402 can include but not limited to a microprocessor MCU or a programmable logic device FPGA processing device), a memory 404 for storing data, and a transmission module 406 for communication function. In addition, it can also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports in the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand that Figure 4 the structure shown is only schematic, and does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 40 can include more or less components than those shown in Figure 4 , or have a different configuration than Figure 4 .

[0126] It should be noted that the one or more processors 402 and / or other data processing circuits described above can be referred to herein as "data processing circuits" in general. The data processing circuit can be embodied in whole or in part as software, hardware, firmware or any other combination. In addition, the data processing circuit can be a single independent processing module, or any one of the other elements combined into the computer terminal 40 in whole or in part. As referred to in the embodiments of the present application, the data processing circuit as a kind of processor control (for example, the selection of the variable resistance terminal path connected with the interface).

[0127] ​The memory 404 can be used to store software programs of application software and modules, such as program instructions / data storage means corresponding to the data processing method in the embodiments of the present application. The processor 402 executes various functional applications and data processing by running the software programs and modules stored in the memory 404, that is, implements the above-mentioned data processing method. The memory 404 can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 404 can further include a memory remotely arranged with respect to the processor 402, which can be connected to the computer terminal 40 through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0128] The transmission module 406 is used to receive or send data via a network. Specific examples of the above-mentioned network can include a wireless network provided by a communication provider of the computer terminal 40. In one example, the transmission module 406 includes a network adapter (NIC), which can be connected to other network devices through a base station so as to be able to communicate with the Internet. In one example, the transmission module 406 can be a radio frequency (RF) module, which is used to communicate with the Internet in a wireless manner.

[0129] The display can be, for example, a touch screen type liquid crystal display (LCD), which can enable a user to interact with the user interface of the computer terminal 40.

[0130] It should be noted that, in some optional embodiments, the above-mentioned Figure 4 The computer terminal shown can include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that, Figure 4 is only one example of a particular implementation and is intended to illustrate the types of components that can be present in the above-mentioned computer terminal.

[0131] It should be noted that, Figure 4 The computer terminal shown is used to execute Figure 1 The data processing method shown, so the related explanations in the execution method of the above-mentioned command also apply to the electronic device, which will not be repeated here.

[0132] The embodiments of the present application also provide a non-volatile storage medium, which includes a stored program, wherein the program controls the device where the storage medium is located to execute the above-mentioned data processing method when running.

[0133] The nonvolatile storage medium executes a program performing the following functions: obtaining live time sequence data corresponding to a video live broadcast, wherein the live time sequence data comprises a plurality of live time node data; classifying live time node data in which a preset condition occurs and live time node data in which the preset condition does not occur among the plurality of live time node data, and determining a live bandwidth demand index according to a classification result, wherein the preset condition comprises that a live quality index of the video live broadcast does not meet a preset requirement due to floating of a network bandwidth index; determining a network quality characteristic quantization factor for representing network performance according to a number of times of network problems occurring in a video live broadcast process; determining a live value characteristic quantization factor for representing live value according to a number of viewers of the video live broadcast and live revenue information; and determining a live acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast according to the live bandwidth demand index, the network quality characteristic quantization factor, and the live value characteristic quantization factor.

[0134] The application also provides an electronic device, comprising a memory and a processor, wherein the processor is configured to run a program stored in the memory, and the program is configured to perform the data processing method.

[0135] The processor is configured to run a program performing the following functions: obtaining live time sequence data corresponding to a video live broadcast, wherein the live time sequence data comprises a plurality of live time node data; classifying live time node data in which a preset condition occurs and live time node data in which the preset condition does not occur among the plurality of live time node data, and determining a live bandwidth demand index according to a classification result, wherein the preset condition comprises that a live quality index of the video live broadcast does not meet a preset requirement due to floating of a network bandwidth index; determining a network quality characteristic quantization factor for representing network performance according to a number of times of network problems occurring in a video live broadcast process; determining a live value characteristic quantization factor for representing live value according to a number of viewers of the video live broadcast and live revenue information; and determining a live acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast according to the live bandwidth demand index, the network quality characteristic quantization factor, and the live value characteristic quantization factor.

[0136] The above sequence numbers of the embodiments of the application are only for description, and do not represent advantages or disadvantages of the embodiments.

[0137] In the above embodiments of the application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0138] In the above embodiments of the present application, the collected information is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with relevant laws, regulations and standards, necessary protection measures are taken, the public order and good customs are not violated, and corresponding operation portals are provided for the user to select authorization or refusal.

[0139] In several embodiments provided in the present application, it should be understood that the disclosed technical content can be implemented by other ways. Among them, the above-described device embodiments are only schematic, for example, the division of the units can be a logical function division, and actual implementation can have another division way, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0140] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place or distributed to multiple units. Part or all of the units can be selected to achieve the purpose of the embodiment scheme according to actual needs.

[0141] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0142] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the part that essentially contributes to the related art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk and various program code storage media.

[0143] The above merely describes the preferred embodiments of the present application, and it should be pointed out that, for those skilled in the art, some improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A data processing method, characterized by, The method comprises the following steps: obtaining live time sequence data corresponding to a video live broadcast, wherein the live time sequence data comprises a plurality of live time node data; classifying live time node data in which a preset condition occurs and live time node data in which the preset condition does not occur among the plurality of live time node data, and determining a live bandwidth demand index according to a classification result, wherein the preset condition comprises a condition in which a live quality index of the video live broadcast does not meet a preset requirement due to floating of a network bandwidth index; determining a network quality characteristic quantification factor for representing network performance according to a number of times that a network problem occurs in the video live broadcast process; determining a live value characteristic quantification factor for representing live value according to a number of viewers of the video live broadcast and live revenue information; determining a live acceleration demand comprehensive quantification evaluation index corresponding to the video live broadcast according to the live bandwidth demand index, the network quality characteristic quantification factor, and the live value characteristic quantification factor; after determining the live acceleration demand comprehensive quantification evaluation index corresponding to the video live broadcast, the method further comprises dynamically adjusting network resource allocation of a live service according to the live acceleration demand comprehensive quantification evaluation index; classifying live time node data in which a preset condition occurs and live time node data in which the preset condition does not occur among the plurality of live time node data, and determining a live bandwidth demand index according to a classification result, comprises: taking a first preset time length as a time window, and counting the live time node data in the time window; assigning a first numerical value to the live time node data in which the preset condition occurs in the time window, and assigning a second numerical value to the live time node data in which the preset condition does not occur in the time window, to obtain a first time sequence corresponding to the time window; obtaining the live time node data in a second preset time length, wherein the second preset time length comprises a plurality of time windows, and performing merging processing on the first time sequences corresponding to the plurality of time windows in the second preset time length to obtain a target time sequence; calculating a mathematical expectation value of the target time sequence, and performing normalization processing on the mathematical expectation value to obtain a network quality level index in a video live broadcast process; taking the first preset time length as an independent variable and the number of the first numerical value as a dependent variable, using a least square method to fit a linear equation, to obtain a first target equation; mapping a first slope in the first target equation to a first digital interval by triangular transformation to obtain a network quality change trend index; and performing weighted summation on the network quality level index and the network quality change trend index to obtain the live bandwidth demand index.

2. The method of claim 1, wherein, determining a network quality characteristic quantification factor for representing network performance according to a number of times that a network problem occurs in the video live broadcast process, comprises: obtaining a number of times that a network problem occurs in a video live broadcast process in a first preset statistical period, wherein the first preset statistical period comprises a plurality of first time steps; determining a first number of times that a network problem occurs in each first time step; add a weight coefficient to a first number corresponding to an i-th first time step in order according to a start time to an end time of the first preset statistical period, to obtain a plurality of second numbers, wherein the weight coefficient is an inverse of a Fibonacci number, a Fibonacci number F[i] corresponding to the i-th first time step is F[i]=F[i-1]+F[i-2], wherein F[0]=1, F[1]=1, and i is a positive integer greater than 1; sum the plurality of second numbers to obtain the network quality characteristic quantization factor.

3. The method of claim 1, wherein, According to the number of viewers of the video live broadcast and the live broadcast income information, a live broadcast value characteristic quantization factor is determined to represent the live broadcast value, including: obtaining the number of viewers of the video live broadcast in a second preset statistical period, wherein the second preset statistical period includes a plurality of second time steps; determining a live broadcast viewer expectation value per unit second time step according to a ratio of the number of viewers to the number of second time steps; generating a second time sequence about the number of viewers corresponding to each second time step and live broadcast income information in each second time step; using a least square regression method to perform linear regression fitting on the second time sequence to obtain a second target equation; mapping a second slope in the second target equation to a second digital interval by a triangular transformation to obtain an income change trend index; multiplying the live broadcast viewer expectation value per unit second time step and the income change trend index to obtain the live broadcast value characteristic quantization factor.

4. The method of claim 1, wherein, determining a live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast, including: obtaining identification information of a target object performing the video live broadcast; determining whether the target object belongs to a multi-channel network organization according to the identification information; in a case where the target object belongs to a multi-channel network organization, multiplying the live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast by a preset weight coefficient to obtain a target live broadcast acceleration demand comprehensive quantization evaluation index.

5. The method of claim 1, wherein, determining a live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast according to the live broadcast bandwidth demand index, the network quality characteristic quantization factor, and the live broadcast value characteristic quantization factor, including: respectively normalizing the live broadcast bandwidth demand index, the network quality characteristic quantization factor, and the live broadcast value characteristic quantization factor, and multiplying the normalized results to obtain the live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast.

6. A data processing apparatus, characterized by, including: an obtaining module configured to obtain live broadcast time sequence data corresponding to a video live broadcast, wherein the live broadcast time sequence data includes a plurality of live broadcast time node data; a first determining module configured to classify live broadcast time node data in which a preset condition occurs and live broadcast time node data in which the preset condition does not occur among the plurality of live broadcast time node data, and determine a live broadcast bandwidth demand index according to a classification result, wherein the preset condition includes a case where a live broadcast quality index of the video live broadcast does not meet a preset requirement due to floating of a network bandwidth index; The second determining module is configured to determine a network quality characteristic quantization factor for representing network performance according to the number of times of network problems occurring in the video live broadcast process; The third determining module is configured to determine a live broadcast value characteristic quantization factor for representing live broadcast value according to the number of viewers of the video live broadcast and live broadcast revenue information; The fourth determining module is configured to determine a live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast according to the live broadcast bandwidth demand index, the network quality characteristic quantization factor, and the live broadcast value characteristic quantization factor. The data processing apparatus is further configured to, after determining the live broadcast acceleration demand comprehensive quantization evaluation index corresponding to the video live broadcast, perform the following step: dynamically adjusting network resource allocation of live broadcast service according to the live broadcast acceleration demand comprehensive quantization evaluation index. The first determining module is further configured to perform the following steps: taking a first preset time length as a time window, counting the live broadcast time node data in the time window; assigning a first numerical value to the live broadcast time node data in the time window that meets the preset condition, and assigning a second numerical value to the live broadcast time node data in the time window that does not meet the preset condition, to obtain a first time sequence corresponding to the time window; obtaining the live broadcast time node data in a second preset time length, wherein the second preset time length includes a plurality of time windows, and performing merging processing on the first time sequences corresponding to the plurality of time windows in the second preset time length to obtain a target time sequence; calculating a mathematical expectation value of the target time sequence, and performing normalization processing on the mathematical expectation value to obtain a network quality level index in a video live broadcast process; using a least square method to fit a linear equation with the first preset time length as an independent variable and the number of the first numerical value as a dependent variable, to obtain a first target equation; mapping a first slope in the first target equation to a first digital interval by triangular transformation to obtain a network quality change trend index; and performing weighted summation on the network quality level index and the network quality change trend index to obtain the live broadcast bandwidth demand index.

7. A non-volatile storage medium, characterized by The non-volatile storage medium includes a stored program, wherein the program controls the device in which the non-volatile storage medium is located to perform the data processing method of any one of claims 1 to 5 when the program is running.

8. A computer program product comprising a computer program, characterized in that, The computer program is executed by a processor to implement the data processing method of any one of claims 1 to 5.

Images