A physical education data processing method and system

Abstract

The application provides a sports education data processing method and system, obtains multiple network indexes of teaching live data at each time point in a previous historical period, and determines network index fluctuation ranges of various network indexes; obtains multiple network indexes of teaching live data at a current live time, and determines an abnormal marker index of the teaching live data at the current live time; when the abnormal marker index is greater than a preset abnormal marker index, a teaching live data response matrix is determined; all deviation value sequences of the teaching live data are determined according to the teaching live data response matrix; an interference coefficient of each deviation value sequence is determined, and then an interference degree of the teaching live data at the current live time is determined according to each deviation value sequence and the interference coefficient of each deviation value sequence; when the interference degree is greater than a preset interference degree, data abnormality early warning is performed on the teaching live data at the current live time, so that picture transmission of the teaching live data is clear and smooth.

Description

Technical Field

[0001] This application relates to the field of sports education data processing technology, and more specifically, to a sports education data processing method and system. Background Technology

[0002] Physical education is undergoing a transformation from traditional to new forms such as integrated online and offline teaching, remote live streaming, and intelligent virtual training. However, physical education differs fundamentally from other subjects. Physical education is "education of the body," emphasizing the demonstration, imitation, and immediate correction of motor skills rather than simply the transmission of knowledge. This necessitates that live streaming physical education instruction possess low latency, high definition, and continuous, stable audio and video transmission capabilities. A close-up of a basketball shot's hand or the complete trajectory of a gymnastic flip requires sufficient image quality and frame rate to ensure students can clearly see the key points of the movement. If the live stream experiences stuttering, audio-visual asynchrony, or pixelation, the details of the movements will be distorted, not only disrupting the teaching rhythm but also potentially leading to misunderstandings and incorrect perceptions of technical movements among students.

[0003] The physical settings of physical education classes are unique. Classes are often conducted in large spaces or semi-outdoor environments such as gymnasiums, athletic fields, and swimming pools. Wi-Fi coverage in these locations is typically uneven, with significant fluctuations in 4G / 5G signal strength. Furthermore, the simultaneous access of numerous smart devices (teacher cameras, student phones, fitness trackers, etc.) can easily lead to channel congestion. Meanwhile, live-streamed teaching places rigid demands on network quality. If network transmission speed is insufficient, data latency is too high, or packet loss rate increases, the encoder will be forced to reduce the bitrate or discard keyframes, directly resulting in blurred images and discontinuous motion. This information gap in teaching caused by network uncertainties has become a core problem that needs to be addressed in current physical education settings. Summary of the Invention

[0004] This application provides a method and system for processing sports education data to solve the technical problem that the data quality of live teaching data is reduced due to various network factors during transmission.

[0005] To solve the above-mentioned technical problems, this application adopts the following technical solution: Firstly, this application provides a method for processing sports education data, including: The teaching live broadcast data platform obtains multiple network indicators of teaching live broadcast data at each time point in the previous historical period, and defines the fluctuation range of each type of network indicator in the historical period to obtain the network indicator fluctuation range of that type of network indicator. Obtain multiple network indicators of the teaching live broadcast data at the current live broadcast time, determine the anomaly markers of each type of network indicator based on the fluctuation range of the network indicator of each type of network indicator, and then determine the anomaly marker index of the teaching live broadcast data at the current live broadcast time based on the anomaly markers of each type of network indicator. When the anomaly labeling index is greater than the preset anomaly labeling index, the multiple network indicators of the teaching live broadcast data at each time point in the previous historical period and the multiple network indicators of the teaching live broadcast data at the current live broadcast time are matrixed to obtain the teaching live broadcast data response matrix. The central reference value is calculated for the column elements of the teaching live data response matrix to obtain a central reference value sequence. The deviation value between the central reference value in the central reference value sequence and each row element of the teaching live data response matrix is ​​calculated to obtain a sequence of all deviation values ​​of the teaching live data. The interference coefficient of each deviation value sequence is determined, and then the interference degree of the teaching live broadcast data at the current live broadcast time is determined based on each deviation value sequence and the interference coefficient of each deviation value sequence. The interference degree is compared with the preset interference degree. When the interference degree is greater than the preset interference degree, a data anomaly warning is issued for the teaching live broadcast data at the current live broadcast time.

[0006] In some embodiments, multiple network metrics include data transmission rate, data latency, and data packet loss rate.

[0007] In some embodiments, the fluctuation range of each type of network metric within the historical period is defined, and the specific fluctuation range of the network metric for that type of network metric includes: Determine the period length of the historical time period; Obtain the expected network index for each type of network index within the historical time period; Determine the volatility of each type of network indicator within the historical period; Determine the fluctuation coefficient of each type of network indicator within the historical period; The upper limit and lower limit of network index fluctuation for each type of network index are determined based on the period length of the historical period, the expected network index of each type of network index within the historical period, the volatility of each type of network index within the historical period, and the volatility coefficient of each type of network index within the historical period. The fluctuation range of a network indicator is determined by the upper limit and lower limit of the fluctuation of each type of network indicator.

[0008] In some embodiments, the upper limit of network metric fluctuation and the lower limit of network metric fluctuation are determined by the following formula:

[0009] in, Indicates the first The upper limit of fluctuation for network-like indicators. Indicates the first The lower limit of network indicator volatility for network-like indicators. Indicates the first [number]th ... Expected network metrics for network metrics Indicates the first [number]th ... The volatility of network-like indicators Indicates the first [number]th ... The volatility coefficient of network-like indicators, Indicates the first [number]th ... Network-like metrics in the first Network metric values ​​at each time point, Indicates the first [number]th ... Network-like metrics in the first Network metric values ​​at each time point, Indicates the first [number]th ... The first type of network metric A time value, Indicates the first [number]th ... The first type of network metric A time value, .

[0010] In some embodiments, determining the anomaly labeling index of the teaching live stream data at the current live streaming moment based on the anomaly labeling of each type of network metric specifically includes: Determine the frequency at which the network metric is marked as an anomaly with a 0-state (the network metric is in a normal state); Determine the frequency at which the network metric is marked as an anomaly (the network metric is in an abnormal state); Determine the total number of states marked with anomalies; The anomaly labeling index of the teaching live broadcast data at the current live broadcast time is determined based on the frequency of the network indicator being marked as 0 (network indicator is in a normal state), the frequency of the network indicator being marked as 1 (network indicator is in an abnormal state), and the total number of anomaly labeling states. The anomaly labeling index is determined by the following formula:

[0011] in, This indicates an anomaly marker index for the current live teaching data. This indicates the total number of exception flags. The abnormal markers for network metrics are: Frequency of states Indicated by Logarithmic function with base 0. This indicates that the exception is marked as state 1 or state 0.

[0012] In some embodiments, before matrixing the various network indicators of the live teaching data at each time point in the previous historical period and the various network indicators of the live teaching data at the current live time, the method further includes standardizing the various network indicators of the live teaching data at each time point in the previous historical period and the various network indicators of the live teaching data at the current live time.

[0013] It should be noted that the preceding historical period should be a time period in which the network status is normal and the teaching live broadcast data does not show any obvious anomalies. This ensures that the defined network indicator fluctuation range accurately reflects the network performance characteristics of the teaching live broadcast data under normal transmission conditions. If the selected historical period itself has network anomalies, the fluctuation range may be too large, thereby reducing the sensitivity of subsequent anomaly detection. In practical applications, periods with normal network status can be identified and selected as historical periods through manual calibration or automatic detection algorithms. Alternatively, multiple historical periods can be used for rolling updates to dynamically adjust the fluctuation range to adapt to the time-varying characteristics of the network environment.

[0014] In some embodiments, calculating the central reference value for the column elements of the teaching live broadcast data response matrix to obtain a central reference value sequence specifically includes: Obtain each column element of the teaching live broadcast data response matrix; Determine the total number of elements in each column of the teaching live broadcast data response matrix; The central reference value of the column element is determined based on each column element of the teaching live broadcast data response matrix and the total number of elements in each column of the teaching live broadcast data response matrix. The center reference values ​​of each column are combined to obtain a sequence of center reference values.

[0015] Secondly, this application provides a sports education data processing system, which includes a network indicator processing unit, the network indicator processing unit comprising: The network indicator fluctuation range determination module is used to obtain multiple network indicators of teaching live broadcast data at each time point in the previous historical period through the teaching live broadcast data platform, and to define the fluctuation range of each type of network indicator in the historical period to obtain the network indicator fluctuation range of that type of network indicator. The anomaly labeling index determination module is used to obtain multiple network indicators of the teaching live broadcast data at the current live broadcast time, determine the anomaly labels of various network indicators of the teaching live broadcast data at the current live broadcast time based on the fluctuation range of each type of network indicator, and then determine the anomaly labeling index of the teaching live broadcast data at the current live broadcast time based on the anomaly labels of each type of network indicator. The teaching live broadcast data response matrix determination module is used to matrixify the multiple network indicators of the teaching live broadcast data at each time point in the previous historical period and the multiple network indicators of the teaching live broadcast data at the current live broadcast time when the anomaly marking index is greater than the preset anomaly marking index, so as to obtain the teaching live broadcast data response matrix. The deviation value sequence determination module is used to calculate the central reference value of the column elements of the teaching live data response matrix to obtain the central reference value sequence, and to calculate the deviation value between the central reference value in the central reference value sequence and each row element of the teaching live data response matrix to obtain all deviation value sequences of the teaching live data. The data anomaly warning module is used to determine the interference coefficient of each deviation value sequence, and then determine the interference degree of the teaching live broadcast data at the current live broadcast time based on each deviation value sequence and the interference coefficient of each deviation value sequence. The interference degree is compared with the preset interference degree. When the interference degree is greater than the preset interference degree, a data anomaly warning is issued for the teaching live broadcast data at the current live broadcast time.

[0016] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described sports education data processing method.

[0017] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described sports education data processing method.

[0018] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The physical education data processing method and system provided in this application first obtains multiple network indicators of teaching live broadcast data at each time point within the previous historical period through a teaching live broadcast data platform, and defines the fluctuation range of each type of network indicator within that historical period to obtain the network indicator fluctuation range of that type of network indicator; then, it obtains multiple network indicators of teaching live broadcast data at the current live broadcast time, determines the anomaly markers of each type of network indicator in the teaching live broadcast data at the current live broadcast time based on the network indicator fluctuation range of each type of network indicator, and then determines the anomaly marker index of the teaching live broadcast data at the current live broadcast time based on the anomaly markers of each type of network indicator; when the anomaly marker index is greater than a preset anomaly marker index, it compares the multiple network indicators of teaching live broadcast data at each time point within the previous historical period with the current live broadcast time... Multiple network indicators of the teaching live broadcast data are matrixed to obtain a teaching live broadcast data response matrix. Central baseline values ​​are calculated for the column elements of the teaching live broadcast data response matrix to obtain a central baseline value sequence. Deviation values ​​are calculated between the central baseline values ​​in the central baseline value sequence and each row element of the teaching live broadcast data response matrix to obtain all deviation value sequences of the teaching live broadcast data. The interference coefficient of each deviation value sequence is determined, and then the interference degree of the teaching live broadcast data at the current live broadcast moment is determined based on each deviation value sequence and its interference coefficient. The interference degree is compared with a preset interference degree. When the interference degree is greater than the preset interference degree, a data anomaly warning is issued for the teaching live broadcast data at the current live broadcast moment, thereby ensuring clear and smooth transmission of the teaching live broadcast data.

[0019] In this application, multiple network indicators of teaching live-stream data at each time point within the previous historical period are obtained through a teaching live-stream data platform, and their fluctuation ranges are defined to obtain the network indicator fluctuation range. This range is used to assess the degree of change in network indicators of the data at the current live-stream moment. When a network indicator is outside the fluctuation range, it means that the network performance during data transmission is severely affected, which will lead to a decrease in the reliability of data transmission. Simultaneously, multiple network indicators of the teaching live-stream data at the current live-stream moment are obtained, and anomaly markers for each type of network indicator are determined based on the fluctuation range. An anomaly marker index is also determined to measure the irregularity and instability of data points at different time points or under different conditions. If the anomaly marker index is greater than a preset value, the network indicators from the previous historical period and the network indicators at the current live-stream moment are merged to form a response. The matrix can effectively analyze whether the influence of various network indicators on the live teaching data at the current moment exceeds the preset range, thereby implementing effective preventive measures. It calculates the central reference value for each column element of the response matrix to obtain a central reference value sequence, and calculates the deviation value sequence between this sequence and each row element of the data response value matrix. It calculates the interference coefficient for each deviation value sequence, and determines the interference level of the live teaching data at the current moment based on the interference coefficient. When the interference level is high, the live teaching data may be subject to significant interference; when the interference level is low, the live teaching data is subject to less interference. It compares the interference level with a preset interference level. If the interference level is greater than the preset value, it issues a data anomaly warning for the live teaching data at the current moment to ensure clear and smooth transmission of the live teaching data. Attached Figure Description

[0020] Figure 1 This is an exemplary flowchart of a physical education data processing method according to some embodiments of this application; Figure 2 This is a flowchart illustrating the process of determining the central reference value sequence in some embodiments of this application; Figure 3 These are schematic diagrams of exemplary hardware and / or software of a network metrics processing unit according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of a computer device for implementing a sports education data processing method according to some embodiments of this application. Detailed Implementation

[0021] The core of this application is to first obtain multiple network indicators of teaching live streaming data at each time point within the previous historical period through a teaching live streaming data platform, and then define the fluctuation range of each type of network indicator within that historical period to obtain the network indicator fluctuation range of that type of network indicator; then obtain multiple network indicators of teaching live streaming data at the current live streaming moment, and determine the anomaly markers of each type of network indicator in the teaching live streaming data at the current live streaming moment based on the network indicator fluctuation range of each type of network indicator, and then determine the anomaly marker index of the teaching live streaming data at the current live streaming moment based on the anomaly markers of each type of network indicator; when the anomaly marker index is greater than a preset anomaly marker index, the multiple network indicators of teaching live streaming data at each time point within the previous historical period and the teaching live streaming data at the current live streaming moment are compared. Multiple network indicators are matrixed to obtain a teaching live broadcast data response matrix. Central baseline values ​​are calculated for the column elements of the teaching live broadcast data response matrix to obtain a central baseline value sequence. Deviation values ​​are calculated between the central baseline values ​​in the central baseline value sequence and each row element of the teaching live broadcast data response matrix to obtain all deviation value sequences of the teaching live broadcast data. The interference coefficient of each deviation value sequence is determined, and then the interference degree of the teaching live broadcast data at the current live broadcast moment is determined based on each deviation value sequence and its interference coefficient. The interference degree is compared with a preset interference degree. When the interference degree is greater than the preset interference degree, a data anomaly warning is issued for the teaching live broadcast data at the current live broadcast moment, thereby ensuring clear and smooth transmission of the teaching live broadcast data.

[0022] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific implementation methods. (Reference) Figure 1 The figure is an exemplary flowchart of a physical education data processing method according to some embodiments of this application. The physical education data processing method 100 mainly includes the following steps: In step 101, multiple network indicators of teaching live data at each time point in the previous historical period are obtained through the teaching live data platform, and the fluctuation range of each type of network indicator in the historical period is defined to obtain the network indicator fluctuation range of that type of network indicator.

[0023] A teaching live streaming data platform is a centralized platform specifically designed for the collection and analysis of live streaming data in physical education teaching. It typically refers to a centralized data center or platform dedicated to collecting, storing, processing, and analyzing large-scale physical education teaching data.

[0024] In some embodiments, multiple network metrics of the live teaching data at each sampling moment within the previous historical period are obtained through a live teaching data platform. That is, each sampling moment of the live teaching data corresponds to a value for each network metric at that moment. In this application, the multiple network metrics include data transmission rate, data latency, and data packet loss rate. Specifically, assuming a basketball shooting lesson where the teacher continuously demonstrates shooting, and network metrics are collected at fixed sampling intervals of 5 seconds, then multiple network metrics of the live teaching data at each sampling moment within the previous historical period are obtained. For example, the multiple network metrics corresponding to the live teaching data at the first sampling moment (5 seconds) are {data transmission rate 1, data latency 1, data packet loss rate 1}; the multiple network metrics corresponding to the second sampling moment (10 seconds) are {data transmission rate 2, data latency 2, data packet loss rate 2}; the multiple network metrics corresponding to the third sampling moment (15 seconds) are {data transmission rate 3, data latency 3, data packet loss rate 3}, and so on, until multiple network metrics for all sampling moments within the previous historical period are obtained. The sampling interval can be set according to the actual teaching scenario, and is not limited here.

[0025] It should be noted that the teaching live data refers to the audio and video teaching content transmitted to students through live streaming during physical education teaching, such as teachers demonstrating basketball shooting, videos explaining gymnastic movements, and demonstrations of running techniques.

[0026] It should be noted that the network indicators refer to some characteristic factors of network performance during data transmission. These network indicators are closely related to the data transmission process. In this application, the network indicators specifically refer to some characteristic factors of network performance during the transmission of live teaching data, such as data transmission rate, data latency, and data packet loss rate. In teaching venues, due to factors such as the large number of mobile phone users and unstable signals, the transmission of live teaching data may be subject to varying degrees of network interference. Therefore, obtaining multiple network indicators of live teaching data can effectively identify the degree of influence on the live teaching data and timely and accurately judge the accuracy and credibility of the current live teaching data.

[0027] In some embodiments, the fluctuation range of each type of network indicator within the historical period is defined. The fluctuation range of the network indicator for this type of network indicator can be obtained in the following way: Determine the period length of the historical time period; Obtain the expected network index for each type of network index within the historical time period; Determine the volatility of each type of network indicator within the historical period; Determine the fluctuation coefficient of each type of network indicator within the historical period; The upper limit and lower limit of network index fluctuation for each type of network index are determined based on the period length of the historical period, the expected network index of each type of network index within the historical period, the volatility of each type of network index within the historical period, and the volatility coefficient of each type of network index within the historical period. The fluctuation range of a network indicator is determined by the upper limit and lower limit of the fluctuation of each type of network indicator.

[0028] In practice, the upper limit and lower limit of network indicator fluctuations are determined by the following formula:

[0029] in, Indicates the first The upper limit of fluctuation for network-like indicators. Indicates the first The lower limit of network indicator volatility for network-like indicators. Indicates the first [number]th ... Expected network metrics for network metrics Indicates the first [number]th ... The volatility of network-like indicators Indicates the first [number]th ... The volatility coefficient of network-like indicators, Indicates the first [number]th ... Network-like metrics in the first Network metric values ​​at each time point, Indicates the first [number]th ... Network-like metrics in the first Network metric values ​​at each time point, Indicates the first [number]th ... The first type of network metric A time value, Indicates the first [number]th ... The first type of network metric A time value, .

[0030] In practice, the period length of the historical time period is determined based on the set historical time period. For example, if half an hour is a period, then the period length is half an hour.

[0031] In specific implementation, the volatility of each type of network indicator within the historical period can be determined by calculating the standard deviation of each type of network indicator within the historical period. That is, the standard deviation of each type of network indicator within the historical period is used as the volatility of that type of network indicator. The volatility of the network indicator refers to the degree of dispersion or volatility of the network indicator within a historical period, representing the degree of dispersion of the network indicator relative to the average network indicator within a historical period. That is, the degree of dispersion of the network indicator. The greater the volatility of the network indicator, the greater the dispersion of the network indicator relative to the average network indicator. The smaller the volatility of the network indicator, the smaller the dispersion of the network indicator relative to the average network indicator.

[0032] In specific implementation, the fluctuation coefficient of each type of network indicator within the historical period is set to a constant between 0 and 1. The fluctuation coefficient reflects the weight of the influence of various network indicators on the quality of teaching live data transmission, and its value can be determined in the following ways: for indicators that directly affect the continuity of the picture (such as data transmission rate), a higher fluctuation coefficient is set; for indicators that have an indirect impact (such as data latency and data packet loss rate), a relatively lower fluctuation coefficient is set. As an optional implementation method, the fluctuation coefficients of data transmission rate, data latency, and data packet loss rate can be set to 0.65, 0.36, and 0.53, respectively. The above values ​​can be obtained through historical data statistical analysis or expert experience calibration, and can be adaptively adjusted according to the network characteristics of the specific teaching scenario in practical applications, which is not limited here. The fluctuation coefficient of the network indicator refers to the degree of influence or fluctuation of the network indicator on the teaching live broadcast data. It can also refer to the instability, irregularity or volatility of the data, that is, the range of change of data points at different times or under different conditions. When the fluctuation coefficient is larger, the influence or fluctuation of the network indicator on the teaching live broadcast data is deeper. When the fluctuation coefficient is smaller, the influence or fluctuation of the network indicator on the teaching live broadcast data is shallower.

[0033] It should be noted that the upper limit of the network indicator fluctuation refers to the highest limit of the network indicator, and the lower limit of the network indicator fluctuation refers to the lowest limit of the network indicator.

[0034] It should be noted that the fluctuation range of the network indicator refers to the range of fluctuation of the network indicator. The fluctuation range of the network indicator is a range defined by the upper limit and lower limit of the fluctuation of the network indicator. Determining the fluctuation range of the network indicator can be used to assess the degree of change of the network indicator of the data at the current live broadcast time. When the network indicator is not within the fluctuation range of the network indicator, it means that the network performance during data transmission is seriously affected, which will lead to a decrease in the reliability of data transmission.

[0035] In step 102, multiple network indicators of the teaching live data at the current live time are obtained. Based on the fluctuation range of each type of network indicator, the anomaly markers of each type of network indicator of the teaching live data at the current live time are determined. Then, the anomaly marker index of the teaching live data at the current live time is determined by the anomaly markers of each type of network indicator.

[0036] In some embodiments, multiple network metrics of the teaching live streaming data at the current live streaming moment are obtained through the teaching live streaming data platform, namely, the data transmission rate, data latency, and data packet loss rate of the teaching live streaming data at the current live streaming moment.

[0037] In some embodiments, determining the anomaly markers for various network metrics in the current live teaching data based on the fluctuation range of each type of network metric can be achieved in the following ways: The various network indicators of the current live teaching data are mapped and compared with the fluctuation range of each type of network indicator. If each type of network indicator in the live teaching data is outside the corresponding fluctuation range at the current live time, the anomaly marker for that type of network indicator is 0; if each type of network indicator in the live teaching data is within the corresponding fluctuation range at the current live time, the anomaly marker for that type of network indicator is 1.

[0038] In practice, the mapping comparison refers to determining whether the various network indicators of the teaching live broadcast data at the current live broadcast time are within the fluctuation range of each type of network indicator.

[0039] It should be noted that the anomaly flag of the network indicator refers to the judgment value used to determine whether the network indicator is within the fluctuation range of the network indicator. The anomaly flag is only represented by the binary values ​​0 and 1. That is, when the anomaly flag of the network indicator is 0, it means that the network indicator is not within the fluctuation range of the network indicator, which means that the network indicator has a significant disturbance and a large degree of uncontrollability. When the anomaly flag of the network indicator is 1, it means that the network indicator is within the fluctuation range of the network indicator, which means that the network indicator has a small disturbance and is within the controllable range.

[0040] In some embodiments, the anomaly labeling index of the teaching live broadcast data at the current live broadcast moment can be determined by the anomaly labeling of each type of network metric in the following manner: Determine the frequency at which the network metric is marked as an anomaly with a 0-state (the network metric is in a normal state); Determine the frequency at which the network metric is marked as an anomaly (the network metric is in an abnormal state); Determine the total number of states marked with anomalies; The anomaly labeling index of the teaching live broadcast data at the current live broadcast time is determined based on the frequency of the network indicator being marked as 0 (network indicator is in a normal state), the frequency of the network indicator being marked as 1 (network indicator is in an abnormal state), and the total number of anomaly labeling states. The anomaly labeling index is determined by the following formula:

[0041] in, This indicates the anomaly marker index for the current live teaching data. This indicates the total number of exception flags. The abnormal markers for network metrics are: Frequency of states Indicated by Logarithmic function with base 0. This indicates that the exception is marked as state 1 or state 0.

[0042] In specific implementation, in this application, the total number of states of the anomaly marker is determined, namely, an anomaly marker of state 0 (network indicator is normal) and an anomaly marker of state 1 (network indicator is abnormal), and the total number of states of the anomaly marker is 2.

[0043] In specific implementation, the frequency of determining the abnormal marking of the network indicator as 0 (network indicator is in normal state) or 1 refers to the number of abnormal markings of multiple network indicators in the current teaching live broadcast data that are 0 and 1.

[0044] It should be noted that the anomaly labeling index refers to a value determined based on the frequency of different anomaly labeling states, used to measure the irregularity and instability of data points at different times or under different conditions. In this application, the anomaly labeling index refers to an irregularity value determined based on the frequency of anomaly labeling state 0 (network indicator is normal) and anomaly labeling state 1 (network indicator is abnormal). The determination of the anomaly labeling index is to identify the degree of influence of multiple network indicators on the teaching live broadcast data at the current live broadcast time. When the anomaly labeling index is larger, the degree of influence of multiple network indicators on the teaching live broadcast data at the current live broadcast time is greater. When the anomaly labeling index is smaller, the degree of influence of multiple network indicators on the teaching live broadcast data at the current live broadcast time is smaller.

[0045] In step 103, when the anomaly labeling index is greater than the preset anomaly labeling index, the multiple network indicators of the teaching live broadcast data at each time point in the previous historical period and the multiple network indicators of the teaching live broadcast data at the current live broadcast time are matrixed to obtain the teaching live broadcast data response matrix.

[0046] In some embodiments, before matrixing the various network indicators of the live teaching data at each time point in the previous historical period and the various network indicators of the live teaching data at the current live time, the method further includes standardizing the various network indicators of the live teaching data at each time point in the previous historical period and the various network indicators of the live teaching data at the current live time, that is, standardizing all network indicators to between 0 and 1, so as to unify the different scales of different network indicators.

[0047] In some embodiments, when the anomaly labeling index is greater than a preset anomaly labeling index, the various network indicators of the teaching live broadcast data at each time point in the previous historical period and the various network indicators of the teaching live broadcast data at the current live broadcast time are matrixed to obtain a teaching live broadcast data response matrix. Specifically, when the anomaly labeling index is greater than the preset anomaly labeling index, the various network indicators of the teaching live broadcast data at each time point in the previous historical period are arranged in rows according to time order, and the various network indicators of the teaching live broadcast data at the current live broadcast time are linked to the last row after the various network indicators of the teaching live broadcast data at each time point in the previous historical period are arranged in rows, thereby obtaining the teaching live broadcast data response matrix.

[0048] It should be noted that the teaching live broadcast data response matrix refers to a matrix obtained by matrix linking multiple network indicators of teaching live broadcast data at each time point in the previous historical period with the multiple network indicators of teaching live broadcast data at the current live broadcast moment. In other words, the teaching live broadcast data response matrix is ​​obtained by combining multiple network indicators from multiple time points in the previous historical period with the multiple network indicators at the current live broadcast moment. It is used to characterize the response of teaching live broadcast data, that is, the response under the influence of multiple network indicators. Since there is a close correlation between teaching live broadcast data transmission and multiple network indicators, the determination of the teaching live broadcast data response matrix can be used to analyze whether the degree of influence of multiple network indicators on the teaching live broadcast data at the current live broadcast moment exceeds the preset range, thereby enabling effective preventive measures.

[0049] In step 104, the central reference value is calculated for the column elements of the teaching live data response matrix to obtain a central reference value sequence. The deviation value between the central reference value in the central reference value sequence and each row element of the teaching live data response matrix is ​​calculated to obtain a sequence of all deviation values ​​of the teaching live data.

[0050] In some embodiments, reference Figure 2 As shown in the figure, this is a flowchart illustrating the process of determining the central reference value sequence in some embodiments of this application. In this embodiment, the determination of the central reference value sequence can be achieved by the following steps: In step 1041, each column element of the teaching live broadcast data response matrix is ​​obtained; Then, in step 1042, the total number of elements in each column of the teaching live data response matrix is ​​determined; Then, in step 1043, the center reference value of the column element is determined based on each column element of the teaching live data response matrix and the total number of each column element of the teaching live data response matrix; Finally, in step 1044, the center reference values ​​of each column element are combined to obtain a center reference value sequence.

[0051] In practice, the central reference value is determined by the following formula:

[0052] in, The teaching live broadcast data response matrix represents the first... The center reference value of the column element, The teaching live broadcast data response matrix represents the first... The total number of column elements, The teaching live broadcast data response matrix represents the first... The first column element One element, .

[0053] It should be noted that the central baseline value sequence refers to a sequence containing multiple central baseline values. The central baseline value is a numerical quantity used to describe the tendency of data in the dataset to the central position. The centrality measure aims to find a single value or numerical value that represents the distribution center of the data as much as possible. In this application, the central baseline value refers to a numerical quantity used to describe the tendency of elements in each column of the teaching live broadcast data response matrix to the central position. Determining the central baseline value can effectively identify the central trend of change of each type of network indicator, thereby more accurately analyzing the impact of teaching live broadcast data.

[0054] In some embodiments, the deviation value between the central reference value in the central reference value sequence and each row element of the teaching live broadcast data response value matrix is ​​calculated to obtain the entire deviation value sequence of the teaching live broadcast data. Specifically, this can be achieved in the following manner: Align the central reference values ​​in the central reference value sequence with each row element of the teaching live broadcast data response value matrix; The residuals are calculated by matching the center reference values ​​in the aligned center reference value sequence with each row element of the teaching live data response value matrix to obtain the sequence of all deviation values ​​of the teaching live data.

[0055] In practice, the residual calculation is the calculation of the difference between two values, which will not be elaborated here.

[0056] It should be noted that the deviation value calculation refers to the process of aligning two sequences of data of the same length and then calculating the residual. In this application, the deviation value calculation of each row element of the central reference value sequence and the teaching live broadcast data response value matrix is ​​to analyze the deviation of each type of network index from the central reference value.

[0057] It should be noted that the deviation value sequence refers to the sequence obtained by calculating the deviation value between the central benchmark value in the central benchmark value sequence and each row element of the teaching live broadcast data response value matrix. The deviation value sequence consists of multiple deviation values. The deviation value is a response value determined by calculating the residual between the central benchmark value in the central benchmark value sequence and each row element of the teaching live broadcast data response value matrix. The determination of the deviation value sequence is based on analyzing the deviation of each type of network indicator from the central benchmark value.

[0058] In step 105, the interference coefficient of each deviation value sequence is determined, and then the interference degree of the teaching live broadcast data at the current live broadcast time is determined based on each deviation value sequence and the interference coefficient of each deviation value sequence. The interference degree is compared with the preset interference degree. When the interference degree is greater than the preset interference degree, a data anomaly warning is issued for the teaching live broadcast data at the current live broadcast time.

[0059] In some embodiments, the interference coefficient of each deviation value sequence can be determined in the following manner: Determine the central tendency of each deviation value sequence; Determine the expected central tendency of each deviation value sequence; The interference coefficient of each deviation value sequence is determined based on its central tendency and its expected central tendency, wherein the interference coefficient is determined by the following formula:

[0060] in, Indicates the first The interference coefficient of a sequence of deviation values. Indicates the first The central tendency of a sequence of deviation values Indicates the first The expected central tendency of a sequence of deviation values.

[0061] In specific implementation, the central tendency of the deviation value sequence can be determined by calculating the average value of the deviation value sequence, that is, the average value of the deviation value sequence is used as the central tendency of the deviation value sequence. In this application, the central tendency of the deviation value sequence refers to the measure of the convergence relationship between different types of network indicators, so as to better analyze the correlation between different types of network indicators.

[0062] In practice, the expected central tendency of each deviation value sequence can be determined through a large number of experiments or machine learning. The expected central tendency refers to the set expected central tendency, which is used to better compare with the current central tendency.

[0063] It should be noted that the interference coefficient of the deviation value sequence refers to the degree of deviation of the data after it has been disturbed.

[0064] In some embodiments, determining the interference level of the live teaching data at the current live broadcast moment based on each deviation value sequence and the interference coefficient of each deviation value sequence can be achieved in the following manner: Obtain the total number of deviation value sequences ; Get the The maximum deviation value in a sequence of deviation values ; Get the The minimum deviation value in a sequence of deviation values ; Get the Interference coefficient of each deviation value sequence ; Get the The first deviation value sequence in the nth deviation value sequence Deviation value ; Determine the first The bullseye value of a sequence of deviation values ; Get the The total number of deviation values ​​in a deviation value sequence ; Based on the total number of the deviation value sequences The first The maximum deviation value in a sequence of deviation values The first The minimum deviation value in a sequence of deviation values The first Interference coefficient of each deviation value sequence The first The first deviation value sequence in the nth deviation value sequence Deviation value The first The bullseye value of a sequence of deviation values and the first The total number of deviation values ​​in a deviation value sequence Determine the interference level of the live teaching data at the current live broadcast moment, wherein the interference level is determined by the following formula:

[0065] in, This indicates the level of interference in the live teaching data at the current moment. This represents the total number of deviation value sequences. Indicates the first The maximum deviation value in a sequence of deviation values. Indicates the first The minimum deviation value in a sequence of deviation values Indicates the first The interference coefficient of a sequence of deviation values. Indicates the first The first deviation value sequence in the nth deviation value sequence One deviation value, Indicates the first The bullseye value of a sequence of deviation values. Indicates the first The total number of deviation values ​​in a sequence of deviation values , .

[0066] In a specific implementation, the bullseye value of the deviation value sequence can be obtained by arranging the deviation values ​​in the deviation value sequence from largest to smallest or smallest to largest, determining the intermediate deviation value from the arranged deviation values, and then obtaining the bullseye value of the deviation value sequence. That is, the intermediate deviation value of the deviation value sequence is taken as the bullseye value of the deviation value sequence. The bullseye value of the deviation value sequence refers to the middle deviation value in the deviation value sequence, and the bullseye value is used to characterize the center position of the deviation values ​​in the deviation value sequence.

[0067] It should be noted that the interference level of the teaching live broadcast data refers to the degree of influence of network indicators during the transmission of teaching live broadcast data. It is an indicator used to measure the irregularity or anomaly of the data. That is, the teaching live broadcast data may be affected by abnormal data transmission rate, data latency, and data packet loss rate during the transmission of teaching live broadcast data. When the teaching live broadcast data is affected by abnormal data transmission rate, data latency, and data packet loss rate, it will seriously affect the efficiency and authenticity of data broadcasting. When the interference level of the teaching live broadcast data is large, the teaching live broadcast data may be subject to greater interference. When the interference level of the teaching live broadcast data is small, the teaching live broadcast data is subject to less interference.

[0068] In some embodiments, the interference level is compared with a preset interference level. When the interference level is greater than the preset interference level, a data anomaly warning is issued for the teaching live broadcast data at the current live broadcast time. Specifically, if the interference level at the current live broadcast time is greater than the preset interference level, it indicates that the current teaching live broadcast data is more irregular or abnormal than normal. In this case, a data anomaly warning will be triggered. The anomaly warning can take various measures, such as notifying the operator, recording the abnormal event, automatically triggering an alarm, or taking further investigation measures.

[0069] It should be noted that the preset interference level is a standard value set based on a large number of experiments or machine learning, used to compare with the current interference level to determine whether the current data is disturbed. The preset interference level is a pre-set threshold that represents the degree of irregularity or abnormality that the data can tolerate under normal circumstances.

[0070] It should be noted that the preset anomaly labeling index and the interference level described in this application can be set based on the statistical distribution characteristics of historical data. Specifically, teaching live broadcast data from multiple historical time periods can be collected, and their anomaly labeling index and interference level can be calculated respectively. The upper quantile (e.g., the 95th percentile) of the calculated value under normal conditions can be used as the corresponding preset threshold. After the threshold is set, it can be dynamically adjusted according to the sensitivity requirements of the actual teaching scenario: if it is desired to improve the sensitivity of anomaly detection, the preset threshold can be appropriately lowered; if it is desired to reduce the false alarm rate, the preset threshold can be appropriately increased. Specific values ​​can be obtained through experimental calibration according to the actual application scenario, and are not limited here.

[0071] Furthermore, in another aspect of this application, in some embodiments, this application provides a sports education data processing system, which includes a network index processing unit, with reference to... Figure 3 The figure is a schematic diagram of exemplary hardware and / or software of a network indicator processing unit according to some embodiments of this application. The network indicator processing unit 200 includes: a network indicator fluctuation range determination module 201, an anomaly marker index determination module 202, a teaching live broadcast data response matrix determination module 203, a deviation value sequence determination module 204, and a data anomaly early warning module 205, which are described below: The network indicator fluctuation range determination module 201 in this application is mainly used to obtain multiple network indicators of teaching live broadcast data at each time point in the previous historical period through the teaching live broadcast data platform, and to define the fluctuation range of each type of network indicator in the historical period to obtain the network indicator fluctuation range of that type of network indicator. The anomaly labeling index determination module 202 in this application is mainly used to obtain multiple network indicators of the teaching live broadcast data at the current live broadcast time, determine the anomaly label of each type of network indicator of the teaching live broadcast data at the current live broadcast time based on the fluctuation range of each type of network indicator, and then determine the anomaly labeling index of the teaching live broadcast data at the current live broadcast time based on the anomaly label of each type of network indicator. The teaching live broadcast data response matrix determination module 203 in this application is mainly used to matrixify the multiple network indicators of the teaching live broadcast data at each time point in the previous historical period and the multiple network indicators of the teaching live broadcast data at the current live broadcast time when the anomaly marking index is greater than the preset anomaly marking index, so as to obtain the teaching live broadcast data response matrix. The deviation value sequence determination module 204 in this application is mainly used to calculate the central reference value of the column elements of the teaching live data response matrix to obtain the central reference value sequence, and to calculate the deviation value between the central reference value in the central reference value sequence and each row element of the teaching live data response matrix to obtain all deviation value sequences of the teaching live data. The data anomaly warning module 205 in this application is mainly used to determine the interference coefficient of each deviation value sequence, and then determine the interference degree of the teaching live broadcast data at the current live broadcast time based on each deviation value sequence and the interference coefficient of each deviation value sequence. The interference degree is compared with the preset interference degree. When the interference degree is greater than the preset interference degree, a data anomaly warning is issued for the teaching live broadcast data at the current live broadcast time.

[0072] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described sports education data processing method.

[0073] In some embodiments, reference Figure 4 The figure is a schematic diagram of the structure of a computer device applying a physical education data processing method according to some embodiments of this application. The physical education data processing method in the above embodiments can... Figure 4 The computer device 300 shown is used to implement this, and the computer device includes at least one processor 301, a communication bus 302, a memory 303 and at least one communication interface 304.

[0074] The processor 301 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (AIC), or one or more for controlling the execution of the sports education data processing method in this application.

[0075] The communication bus 302 may include a path for transmitting information between the aforementioned components.

[0076] The memory 303 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or it may be an electrically erasable programmable read-only memory (EEPROM), a compact dic read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), a magnetic disk or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. The memory 303 may exist independently and be connected to the processor 301 via a communication bus 302. The memory 303 may also be integrated with the processor 301.

[0077] The memory 303 stores program code that executes the scheme of this application, and its execution is controlled by the processor 301. The processor 301 executes the program code stored in the memory 303. The program code may include one or more software modules. In the above embodiments, the determination of the anomaly marker index can be achieved by the processor 301 and one or more software modules in the program code in the memory 303.

[0078] Communication interface 304 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.

[0079] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (independent CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0080] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital device (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0081] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described sports education data processing method.

[0082] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0083] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of the invention. Therefore, if these modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A method for processing physical education data, characterized in that, Includes the following steps: The teaching live broadcast data platform obtains multiple network indicators of teaching live broadcast data at each time point in the previous historical period, and defines the fluctuation range of each type of network indicator in the historical period to obtain the network indicator fluctuation range of that type of network indicator. Obtain multiple network indicators of the teaching live broadcast data at the current live broadcast time, determine the anomaly markers of each type of network indicator based on the fluctuation range of the network indicator of each type of network indicator, and then determine the anomaly marker index of the teaching live broadcast data at the current live broadcast time based on the anomaly markers of each type of network indicator. When the anomaly labeling index is greater than the preset anomaly labeling index, the multiple network indicators of the teaching live broadcast data at each time point in the previous historical period and the multiple network indicators of the teaching live broadcast data at the current live broadcast time are matrixed to obtain the teaching live broadcast data response matrix. The central reference value is calculated for the column elements of the teaching live data response matrix to obtain a central reference value sequence. The deviation value between the central reference value in the central reference value sequence and each row element of the teaching live data response matrix is ​​calculated to obtain a sequence of all deviation values ​​of the teaching live data. The interference coefficient of each deviation value sequence is determined, and then the interference degree of the teaching live broadcast data at the current live broadcast time is determined based on each deviation value sequence and the interference coefficient of each deviation value sequence. The interference degree is compared with the preset interference degree. When the interference degree is greater than the preset interference degree, a data anomaly warning is issued for the teaching live broadcast data at the current live broadcast time.

2. The method as described in claim 1, characterized in that, Several network metrics include data transmission rate, data latency, and data packet loss rate.

3. The method as described in claim 1, characterized in that, The fluctuation range of each type of network indicator within this historical period is defined, and the specific fluctuation range of that type of network indicator includes: Determine the period length of the historical time period; Obtain the expected network index for each type of network index within the historical time period; Determine the volatility of each type of network indicator within the historical period; Determine the fluctuation coefficient of each type of network indicator within the historical period; The upper limit and lower limit of network index fluctuation for each type of network index are determined based on the period length of the historical period, the expected network index of each type of network index within the historical period, the volatility of each type of network index within the historical period, and the volatility coefficient of each type of network index within the historical period. The fluctuation range of a network indicator is determined by the upper limit and lower limit of the fluctuation of each type of network indicator.

4. The method as described in claim 3, characterized in that, The upper limit and lower limit of the network indicator fluctuation are determined by the following formula: in, Indicates the first The upper limit of fluctuation for network-like indicators. Indicates the first The lower limit of network indicator volatility for network-like indicators. Indicates the first [number]th ... Expected network metrics for network metrics Indicates the first [number]th ... The volatility of network-like indicators Indicates the first [number]th ... The volatility coefficient of network-like indicators, Indicates the first [number]th ... Network-like metrics in the first Network metric values ​​at each time point, Indicates the first [number]th ... Network-like metrics in the first Network metric values ​​at each time point, Indicates the first [number]th ... The first type of network metric A time value, Indicates the first [number]th ... The first type of network metric A time value, .

5. The method as described in claim 1, characterized in that, The anomaly labeling index for the current live teaching data is determined by the anomaly labeling of each type of network metric. Specifically, it includes: Determine the frequency at which the network metric is marked as an anomaly with a state of 0; Determine the frequency at which the network metric is marked as an anomaly (1 state); Determine the total number of states marked with anomalies; The anomaly labeling index of the teaching live broadcast data at the current live broadcast time is determined based on the frequency of the network metric being marked as 0, the frequency of the network metric being marked as 1, and the total number of anomaly labeling states. The anomaly labeling index is determined by the following formula: in, This indicates an anomaly marker index for the current live teaching data. This indicates the total number of exception flags. The abnormal markers for network metrics are: Frequency of states Indicated by Logarithmic function with base 0. This indicates that the exception is marked as state 1 or state 0.

6. The method as described in claim 1, characterized in that, Before matrixing the various network indicators of the live teaching data at each time point in the previous historical period and the various network indicators of the live teaching data at the current live time, it is also necessary to standardize the various network indicators of the live teaching data at each time point in the previous historical period and the various network indicators of the live teaching data at the current live time.

7. The method as described in claim 1, characterized in that, The central reference value is calculated for each column element of the teaching live broadcast data response matrix to obtain the central reference value sequence, which specifically includes: Obtain each column element of the teaching live broadcast data response matrix; Determine the total number of elements in each column of the teaching live broadcast data response matrix; The central reference value of the column element is determined based on each column element of the teaching live broadcast data response matrix and the total number of elements in each column of the teaching live broadcast data response matrix. The center reference values ​​of each column are combined to obtain a sequence of center reference values.

8. A sports education data processing system, characterized in that, It includes a network metric processing unit, which includes: The network indicator fluctuation range determination module is used to obtain multiple network indicators of teaching live broadcast data at each time point in the previous historical period through the teaching live broadcast data platform, and to define the fluctuation range of each type of network indicator in the historical period to obtain the network indicator fluctuation range of that type of network indicator. The anomaly labeling index determination module is used to obtain multiple network indicators of the teaching live broadcast data at the current live broadcast time, determine the anomaly labels of various network indicators of the teaching live broadcast data at the current live broadcast time based on the fluctuation range of each type of network indicator, and then determine the anomaly labeling index of the teaching live broadcast data at the current live broadcast time based on the anomaly labels of each type of network indicator. The teaching live broadcast data response matrix determination module is used to matrixify the multiple network indicators of the teaching live broadcast data at each time point in the previous historical period and the multiple network indicators of the teaching live broadcast data at the current live broadcast time when the anomaly marking index is greater than the preset anomaly marking index, so as to obtain the teaching live broadcast data response matrix. The deviation value sequence determination module is used to calculate the central reference value of the column elements of the teaching live data response matrix to obtain the central reference value sequence, and to calculate the deviation value between the central reference value in the central reference value sequence and each row element of the teaching live data response matrix to obtain all deviation value sequences of the teaching live data. The data anomaly warning module is used to determine the interference coefficient of each deviation value sequence, and then determine the interference degree of the teaching live broadcast data at the current live broadcast time based on each deviation value sequence and the interference coefficient of each deviation value sequence. The interference degree is compared with the preset interference degree. When the interference degree is greater than the preset interference degree, a data anomaly warning is issued for the teaching live broadcast data at the current live broadcast time.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing code, and the processor being configured to retrieve the code and execute the sports education data processing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the sports education data processing method as described in any one of claims 1 to 7.

Images