A method, device, medium and computer device for evaluating playing quality

Abstract

The application provides a method, device, medium and computer equipment for evaluating playing quality, the method comprising: acquiring playing failure data and playing lag data; determining a playing failure rate based on the playing failure data; determining a playing lag rate based on the playing lag data; determining a quality evaluation score based on the playing failure rate and the playing lag rate; and evaluating playing quality by using the quality evaluation score. Thus, the application uses the collected data of a playing terminal as a data source to obtain streaming media playing quality evaluation data (playing lag data and playing failure data), and the proportion of user subjective feedback in the data is very small, and the proportion of objective parameters is large, so that the evaluation quality of the streaming media playing can be ensured, and the quality evaluation precision is improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of Internet live broadcast, and particularly relates to a method and device for evaluating playing quality, a medium and a computer device. BACKGROUND

[0002] For streaming services on the Internet, quantitatively evaluating playing quality is of great help and significance to user experience and improvement of streaming service quality.

[0003] In related technologies, the evaluation method of playing quality is too closely combined with the scene, and user subjective feedback accounts for a large proportion, is affected by many subjective factors, and is not strong in universality, so that the quality evaluation precision cannot be ensured. SUMMARY

[0004] In view of the problems in the prior art, the embodiments of the present application provide a method and device for evaluating playing quality, a medium and a computer device, to solve the technical problem that the quality evaluation precision is not high and cannot be ensured when evaluating playing quality in the prior art.

[0005] The present application provides a method for evaluating playing quality, comprising:

[0006] obtaining playing failure data and playing lag data;

[0007] determining a playing failure rate based on the playing failure data;

[0008] determining a playing lag rate based on the playing lag data;

[0009] determining a quality evaluation score based on the playing failure rate and the playing lag rate;

[0010] evaluating playing quality by using the quality evaluation score.

[0011] Optionally, the obtaining playing data comprises:

[0012] obtaining the number of successfully played videos, the number of failed videos and the first total amount of video collection in a preset collection time section.

[0013] Optionally, the obtaining playing data comprises:

[0014] obtaining the number of lagging videos, the lag time and the second total amount of video collection in a preset collection time section.

[0015] Optionally, the determining a playing failure rate based on the playing failure data comprises:

[0016] according to the formula determining the play failure rate Playrate; wherein the Sf i is the number of videos that fail to play in the i-th collection time section; the St i is the corresponding first total amount of video collection in the i-th collection time section; the N is the total number of collection time sections, and the i is the serial number of each collection time section.

[0017] Optionally, the play stall rate is determined based on the play stall data, including:

[0018] According to the formula determining the play stall rate Kdrate; wherein Kd i is the number of videos that stall in the i-th collection time section; the Ht i is the corresponding second total amount of video collection in the i-th collection time section, the Kdt i is the corresponding stall duration of the i-th collection time section, the Time is the collection duration corresponding to the i-th collection time section, the N is the total number of collection time sections, and the i is the serial number of each collection time section.

[0019] Optionally, the quality evaluation score is determined based on the play failure rate and the play stall rate, including:

[0020] According to the formula determining the quality evaluation score Q; wherein the w i is the stall weight corresponding to the i-th collection time section, Kdrate i is the stall rate corresponding to the i-th collection time section, the m i is the play failure weight corresponding to the i-th collection time section, Playdrate i is the play failure rate corresponding to the i-th collection time section, the Deno is a scoring constant, the N is the total number of collection time sections, and the i is the serial number of each collection time section.

[0021] Optionally, the play quality is evaluated by using the quality evaluation score, including:

[0022] determining the score interval range to which the quality evaluation score belongs;

[0023] finding the corresponding quality level from a mapping table according to the score interval range; the mapping table has pre-stored corresponding relationship between each score interval range and the quality level.

[0024] The application further provides a device for evaluating play quality, the device comprising:

[0025] An acquisition unit is configured to acquire playback failure data and playback lag data.

[0026] A determination unit is configured to determine a playback failure rate based on the playback failure data, determine a playback lag rate based on the playback lag data, and determine a quality evaluation score based on the playback failure rate and the playback lag rate.

[0027] An evaluation unit is configured to evaluate the playback quality by using the quality evaluation score.

[0028] The application further provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method of any one of the above.

[0029] The application further provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method of any one of the above when executing the program.

[0030] The application provides a method, device, medium, and computer device for evaluating playback quality, the method comprising: acquiring playback failure data and playback lag data; determining a playback failure rate based on the playback failure data; determining a playback lag rate based on the playback lag data; determining a quality evaluation score based on the playback failure rate and the playback lag rate; and evaluating the playback quality by using the quality evaluation score. BRIEF DESCRIPTION OF DRAWINGS

[0031] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments, and are not intended to limit the scope of the application. Moreover, the same reference numerals are used throughout the drawings to represent similar components. In the drawings:

[0032] Figure 1 A method flowchart for evaluating playback quality is provided for the embodiments of the application.

[0033] Figure 2 A device structure diagram for evaluating playback quality is provided for the embodiments of the application.

[0034] Figure 3 A computer device structure diagram for evaluating playback quality is provided for the embodiments of the application.

[0035] Figure 4The computer storage medium structure schematic diagram for evaluating the playing quality provided by the embodiment of the present application. DETAILED DESCRIPTION

[0036] In order to solve the technical problem that the quality evaluation precision cannot be ensured in the prior art when evaluating the playing quality, the present application provides a method, device, medium and computer equipment for evaluating the playing quality.

[0037] In order to better understand the above technical solution, the technical solution of the present application will be described in detail below by means of the accompanying drawings and specific embodiments. It should be understood that the specific features in the present application and the embodiments are detailed descriptions of the technical solution of the present application, and are not limitations of the technical solution of the present application. In the case of no conflict, the technical features in the present application and the embodiments can be combined with each other.

[0038] The present embodiment provides a method for evaluating the playing quality, as shown in the method comprises: Figure 1

[0039] S110, obtaining playing failure data and playing lag data;

[0040] Due to network or cache reasons, the cache data to be rendered in the memory is not enough, and the next frame rendering cannot be completed. This situation is called player playing lag.

[0041] When the player plays streaming media data, if the network or memory causes the player to black out, crash, network connection to be disconnected, and other situations that affect user viewing are called playing failure.

[0042] This step can obtain the playing failure data and playing lag data of each user when the player plays streaming media data, so as to evaluate the streaming media playing quality.

[0043] Here, obtaining the playing data comprises:

[0044] In a preset collection time section, the number of successfully played videos, the number of failed videos and the first video collection total amount are obtained.

[0045] For example, if the collection time section is 1 hour, then the number of successfully played videos, the number of failed videos and the first video collection total amount in each 1 hour are collected.

[0046] ​The embodiment includes two playing parameters: a playing success parameter Sc and a playing failure parameter Sf. For example, if four playing data are collected in the first collection time section (i=1), the number of videos with playing success is two, and the number of videos with playing failure is two, then Sf1=2 and Sc1=2; if four playing data are collected in the second collection time section (i=2), the number of videos with playing success is three, and the number of videos with playing failure is one, then Sf1=1 and Sc1=2.

[0047] Similarly, playing stall data can also be obtained in the playing process. The obtained playing data include:

[0048] In the preset collection time section, the number of videos with playing stall, the stall duration and the second video collection total amount are obtained.

[0049] For example, if the collection time section is one hour, then the number of videos with playing stall and the second video collection total amount in each one hour are obtained.

[0050] The embodiment includes two stall parameters: the number of videos with playing stall Kd and the stall duration Kdt. In the playing process, for example, two videos are collected in a collection time section. If one video has stall, then Kd=1, the stall duration is Kdt, and the second video collection total amount Ht=2.

[0051] It is worth noting that the collection of playing failure / success data and playing stall data is two independent collection processes. For the same collection time section, playing failure / success data and playing stall data can be collected at the same time.

[0052] In this step, the data related to playing quality are obtained as the main data for evaluating playing quality, without excessive dependence on subjective evaluation of users, so as to ensure the objectivity of evaluation factors and the accuracy of evaluation quality.

[0053] In S111, a playing failure rate is determined based on the playing failure data.

[0054] After the playing failure data are obtained, the playing failure rate is determined based on the playing failure data.

[0055] In the embodiment, the playing failure rate is determined based on the playing failure data, including:

[0056] According to the formula the playing failure rate Playrate is determined; wherein Sf i is the number of videos with playing failure in the ith collection time section; St iis the first video collection total amount corresponding to the i-th collection time section; N is the total number of collection time sections, and i is the serial number of each collection time section.

[0057] The principle of the above formula is that the formula is based on the collection of large-scale playback data, and the quotient value between the number of videos with playback failure and the first video collection total amount is taken as the playback failure rate.

[0058] For example, continuing the above example, if 4 playback data are collected in the first collection time section (i = 1), of which the number of videos with successful playback is 2 and the number of videos with playback failure is 2, then Sf1 = 2 and St1 = 4; if 4 playback data are collected in the second collection time section (i = 2), of which the number of videos with successful playback is 2 and the number of videos with playback failure is 1, then Sf1 = 1 and St1 = 4.

[0059] Then

[0060] This step can quantitatively evaluate the playback failure rate of the client by counting the number of playback failures and the total number of playbacks, and then determining the playback failure rate by using the number of actual playback reports and the total number of playbacks, thereby reducing the proportion of user subjective feedback and improving the evaluation quality of the streaming media playback; and the more sample data collected, the closer to the real situation of real users, and the higher the evaluation accuracy.

[0061] S112, determining a playback freezing rate based on the playback freezing data;

[0062] Similarly, after obtaining the playback freezing data, the playback failure rate is determined based on the playback freezing data.

[0063] In this embodiment, determining the playback freezing rate based on the playback freezing data comprises:

[0064] According to the formula to determine the playback freezing rate Kdrate; wherein Kd i is the number of videos with playback freezing in the i-th collection time section; Ht i is the second video collection total amount corresponding to the i-th collection time section, Kdt i is the freezing duration corresponding to the i-th collection time section, Time is the collection duration corresponding to the i-th collection time section, N is the total number of collection time sections, and i is the serial number of each collection time section.

[0065] It is worth noting that the collection duration Time corresponding to each collection time section can be the same or different.

[0066] The principle of the above formula is: the playback stall rate is determined by using the stall times, the stall duration and the second video collection total amount, and because the time interval of collection is uneven, the Time factor is considered to ensure the stability of the overall evaluation.

[0067] For example: if 2 playback data are collected in the first collection time section (i=1), the collection duration is 60s, among which the number of videos with playback stall is 1 and the stall duration is 10s, then Kd1=1, Kdt1=10 and Ht1=2; if 3 playback data are collected in the second collection time section (i=2), the collection duration is 60s, the number of videos with playback stall is 2 and the stall duration is 10s, then Kd2=1, Kdt2=10 and Ht2=3; Time=60; then,

[0068]

[0069] This step determines the playback stall rate by using the stall times, the stall duration and the second video collection total amount, and because the stall times and the stall duration are objective factors affecting the video playback quality and the proportion of user subjective feedback is small, the evaluation accuracy can be ensured; and because the collection time section can be uneven, the Time factor is considered to ensure the stability of the overall evaluation.

[0070] S113, determining a quality evaluation score based on the playback failure rate and the playback stall rate;

[0071] After the playback failure rate and the playback stall rate are determined, a quality evaluation score is determined based on the playback failure rate and the playback stall rate.

[0072] In this embodiment, the quality evaluation score is determined based on the playback failure rate and the playback stall rate, including:

[0073] According to the formula to determine the quality evaluation score Q; wherein w i Kdrate i is the stall rate corresponding to the ith collection time section, m i Playdrate i is the playback failure rate corresponding to the ith collection time section, Deno is a scoring constant, N is the total number of collection time sections, and i is the serial number of each collection time section.

[0074] Principle of the above formula: the quality evaluation score of the video is determined for each playback stall rate and playback failure rate in each collection time segment; and because the time segments are divided into business hot time segments and non-business hot time segments, a weight is preset for the playback stall rate and the playback failure rate of each time segment, so that Q is closer to the actual situation, thereby ensuring the accuracy of Q.

[0075] Because the playback quality is different in the business hot time segment and the non-business hot time segment, in order to improve the evaluation accuracy, the playback failure weight and the stall weight corresponding to different time segments are different in the embodiment.

[0076] For example, such as from 8 am to 6 pm is a non-business hot time segment, and from 8 pm to 1 am is a business hot time segment. If 1 hour is divided, the business time segment can be divided into 15 collection time segments, the non-business hot time segment includes 10 collection time segments, and the business hot time segment includes 5 collection time segments, and the serial number of each collection time segment starts from 1 and increases. Then the playback failure weight and the stall weight corresponding to the business hot time segment can be set to be larger, and the playback failure weight and the stall weight corresponding to the non-business hot time segment can be set to be smaller.

[0077] Specifically, the playback failure weight and the stall weight are both 10-100, the value of the playback failure weight and the stall weight corresponding to the non-business hot time segment can be in the range of 10-50, and the value of the playback failure weight and the stall weight corresponding to the business hot time segment can be in the range of 60-100.

[0078] It is worth noting that the value of the playback failure weight and the stall weight is not fixed in the non-business hot time segment or the business hot time segment, but needs to be continuously increased with the serial number of the collection time segment. For example, for the playback failure weight, for example, from 8 pm to 12 pm, the playback failure weight corresponding to 8 pm-9 pm can be 60, the playback failure weight corresponding to 9 pm-10 pm can be 70, and the playback failure weight corresponding to 10 pm-10 pm can be 80, and so on.

[0079] Deno is a scoring constant, and the value of Deno needs to be determined based on a preset convergence interval, and the final quality evaluation score needs to be converged in the preset convergence interval.

[0080] For example, if the current quality evaluation score obtained is 449, and the convergence interval is 0-100, the specific value of the scoring constant can be determined based on the current quality evaluation score and the convergence interval.

[0081] For example, assume that there are four collection time sections, and the corresponding stalling weight, stalling rate, playing failure rate and playing failure weight of each collection time section are as follows:

[0082] Kdrate1=0.1,PlaydRate1=0.1,w1=0.25,m1=0.25

[0083] Kdrate2=0.2,PlaydRate2=0.2,w2=0.25,m2=0.25

[0084] Kdrate3=0.3,PlaydRate3=0.3,w3=0.25,m3=0.25

[0085] Kdrate4=0.2,PlaydRate4=0.2,w4=0.25,m4=0.25

[0086] Then Q=(0.1*0.25+0.2*0.25+0.3*0.25+0.2*0.25+0.1*0.25+0.2*0.25+0.3*0.25+0.2*0.25) / 4*Deno

[0087] =(0.2+0.2) / 4*Deno

[0088] =0.1*(Deno=10000)

[0089] =1000

[0090] It is worth noting that when Deno is 10000, the evaluation score can be converged into the preset convergence interval.

[0091] In this step, the playing failure weight corresponding to each collection time section is different when calculating the quality evaluation score, and the stalling weight corresponding to each collection time section is also different, which fully considers the actual playing situation and thus ensures the playing quality evaluation accuracy.

[0092] S114, evaluating the playing quality by using the quality evaluation score.

[0093] After determining the quality evaluation score, the playing quality is evaluated by using the quality evaluation score.

[0094] Evaluating the playing quality by using the quality evaluation score, comprising:

[0095] Determining the score interval range to which the quality evaluation score belongs;

[0096] Looking up the corresponding quality level from the mapping table according to the score interval range, wherein the mapping table pre-stores the corresponding relationship between each score interval range and quality level.

[0097] Specifically, the score interval range includes: [1200, 1100), [1100, 900), [900, 700), [700, 400), [400, 0); the quality level corresponding to [1200, 1100) is excellent, the quality level corresponding to [1100, 900) is excellent and good, the quality level corresponding to [900, 700) is general, the quality level corresponding to [700, 400) is poor, and the quality level corresponding to [400, 0) is very poor.

[0098] Then, if it is determined that the quality evaluation score belongs to the interval range [1200, 1100), it is determined that the playing quality is excellent; if it is determined that the quality evaluation score belongs to the interval range [1100, 900), it is determined that the playing quality is good; if it is determined that the quality evaluation score belongs to the interval range [900, 700), it is determined that the playing quality is general; if it is determined that the quality evaluation score belongs to the interval range [700, 400), it is determined that the playing quality is poor; and if it is determined that the quality evaluation score belongs to the interval range [400, 0), it is determined that the playing quality is very poor.

[0099] As described in the above example, when Q is 1000, it is determined that the quality evaluation score Q belongs to the interval range [1100, 900), and it is determined that the playing quality of the live broadcast platform is good.

[0100] The method for evaluating playing quality provided in the embodiment uses the collection data of the playing end as a data source to obtain the streaming media playing quality evaluation data (playing lag data and playing failure data), and the subjective feedback of the user accounts for a very small proportion in the data, and the objective parameters account for a large proportion, so that the evaluation quality of the streaming media playing can be ensured, and the quality evaluation precision is improved.

[0101] Based on the same inventive concept, the embodiment also provides a device for evaluating playing quality, as shown in Figure 2 The device includes:

[0102] The acquisition unit 21 is configured to acquire playing failure data and playing lag data.

[0103] The determination unit 22 is configured to determine a playing failure rate based on the playing failure data, determine a playing lag rate based on the playing lag data, and determine a quality evaluation score based on the playing failure rate and the playing lag rate.

[0104] The evaluation unit 23 is configured to evaluate the playing quality by using the quality evaluation score.

[0105] Specifically, due to network or cache reasons, the cache data to be rendered in the memory is not enough, and the next frame rendering cannot be completed. This situation is called player playback stuttering.

[0106] When the player plays streaming media data, if the player black screen, crash, network disconnection and other situations affect the user's viewing are caused by network or memory reasons, it is called playback failure.

[0107] In this embodiment, when playing streaming media data using the player, the playback failure data and playback stuttering data of each user can be obtained by the acquisition unit 21 to evaluate the streaming media playback quality.

[0108] The acquisition unit 21 acquires the playback data, which includes:

[0109] In the preset collection time section, the number of successfully played videos, the number of failed videos, and the first video collection total are obtained.

[0110] For example, if the collection time section is 1 hour, then the number of successfully played videos, the number of failed videos, and the first video collection total in each 1 hour are collected.

[0111] This embodiment includes two playback parameters: a successful playback parameter Sc and a failed playback parameter Sf. For example, if 4 playback data are collected in the first collection time section (i=1), the number of successfully played videos is 2, and the number of failed videos is 2, then Sf1=2 and Sc1=2; if 4 playback data are collected in the second collection time section (i=2), the number of successfully played videos is 3, and the number of failed videos is 1, then Sf1=1 and Sc1=2.

[0112] Similarly, during the playback process, playback stuttering data can also be obtained. Therefore, the acquisition unit 21 acquires the playback data, which includes:

[0113] In the preset collection time section, the number of stuttered videos, the stuttering duration, and the second video collection total are obtained.

[0114] For example, if the collection time section is 1 hour, then the number of stuttered videos and the second video collection total in each 1 hour are collected.

[0115] This embodiment includes two stuttering parameters: the number of stuttered videos Kd and the stuttering duration Kdt; during the playback process, for example, in a certain collection time section, two videos are collected, if one of the videos stutters, then Kd=1, the stuttering duration is Kdt, and the second video collection total Ht=2.

[0116] It is worth noting that the collection of play failure / success data and the collection of play stutter data are two independent collection processes. For the same collection time segment, play failure / success data and play stutter data can be collected simultaneously.

[0117] In this step, the data related to the play quality is obtained as the main data for evaluating the play quality, without excessive dependence on the subjective evaluation of the user, ensuring the objectivity of the evaluation factor and the accuracy of the evaluation quality.

[0118] After obtaining the play failure data, the determination unit 22 determines the play failure rate based on the play failure data.

[0119] In this embodiment, the determination unit 22 determines the play failure rate based on the play failure data, including:

[0120] According to the formula determines the play failure rate Playrate; wherein Sf i is the number of videos that fail to play in the i-th collection time segment; St i is the corresponding first video collection total in the i-th collection time segment; N is the total number of collection time segments, and i is the serial number of each collection time segment.

[0121] The principle of the above formula is that based on the collection of large-scale play data, the quotient value between the number of videos that fail to play and the first video collection total is taken as the play failure rate.

[0122] For example, continuing the above example, if in the first collection time segment (i = 1), 4 play data are collected, of which the number of videos that play successfully is 2 and the number of videos that fail to play is 2, then Sf1 = 2 and St1 = 4; if in the second collection time segment (i = 2), 4 play data are collected, of which the number of videos that play successfully is 2 and the number of videos that fail to play is 1, then Sf1 = 1 and St1 = 4.

[0123] Then

[0124] In this step, the number of play failures and the total number of plays are counted, and then the number of actual play reports and the total number of plays are used to determine the play failure rate. The play failure rate of the client can be quantitatively evaluated, the proportion of user subjective feedback is reduced, and the evaluation quality of the streaming media play is improved. Moreover, the more sample data collected, the closer to the real situation of real users, and the higher the evaluation accuracy.

[0125] Similarly, after obtaining the play stutter data, the determination unit 22 determines the play failure rate based on the play stutter data.

[0126] In this embodiment, the determination unit 22 determines the playback stall rate based on the playback stall data, including:

[0127] According to the formula determines the playback stall rate Kdrate; wherein, Kd i is the number of videos with playback stall in the i-th collection time segment; Ht i is the second video collection total amount corresponding to the i-th collection time segment, Kdt i is the stall duration corresponding to the i-th collection time segment, Time is the collection duration corresponding to the i-th collection time segment, N is the total number of collection time segments, and i is the serial number of each collection time segment.

[0128] It is worth noting that the corresponding collection duration Time in each collection time segment can be the same or different.

[0129] The principle of the above formula is: the playback stall rate is determined by using the stall frequency, stall duration, and second video collection total amount, and because the time interval of collection is uneven, the Time factor is considered to ensure the stability of the overall evaluation.

[0130] For example: if in the first collection time segment (i=1), 2 playback data are collected, the collection duration is 60s, among which the number of videos with playback stall is 1 and the stall duration is 10s, then Kd1=1, Kdt1=10, Ht1=2; if in the second collection time segment (i=2), 3 playback data are collected, the collection duration is 60s, the number of videos with playback stall is 2, and the stall duration is 10s, then Kd2=1, Kdt2=10, Ht2=3; Time=60; then,

[0131]

[0132] This step determines the playback stall rate by using the stall frequency, stall duration, and second video collection total amount. Because the stall frequency and stall duration are objective factors affecting video playback quality, the proportion of user subjective feedback is small, so the evaluation accuracy can be ensured; and because the time interval of collection is uneven, the Time factor is considered to ensure the stability of the overall evaluation.

[0133] After the playback failure rate and the playback stall rate are determined, the determination unit 22 determines the quality evaluation score based on the playback failure rate and the playback stall rate.

[0134] In this embodiment, the determination unit 22 determines the quality evaluation score based on the playback failure rate and the playback stall rate, including:

[0135] According to the formula determines the quality evaluation score Q; wherein, wi Kdrate is the stalling weight corresponding to the i-th collection time section i m is the stalling rate corresponding to the i-th collection time section i Playdrate is the playing failure weight corresponding to the i-th collection time section i Deno is the playing failure rate corresponding to the i-th collection time section, N is the total number of collection time sections, and i is the serial number of each collection time section.

[0136] The principle of the above formula is that the quality evaluation score of the video is determined according to the playing stalling rate and the playing failure rate in each collection time section; and because the time section is divided into a business hot time section and a non-business hot time section, a weight is preset for the playing stalling rate and the playing failure rate of each time section, so that Q is closer to the actual situation, thereby ensuring the accuracy of Q.

[0137] Because the playing quality is different in the business hot time section and the non-business hot time section, in order to improve the evaluation accuracy, the playing failure weight and the stalling weight corresponding to different time sections are different in this embodiment.

[0138] For example, the non-business hot time section is from 8 am to 6 pm, and the business hot time section is from 8 pm to 1 am. If the business time section is divided into 15 collection time sections at an interval of 1 hour, the non-business hot time section includes 10 collection time sections, and the business hot time section includes 5 collection time sections, and the serial number of each collection time section starts from 1 and increases. Then the playing failure weight and the stalling weight corresponding to the business hot time section are set to be larger, and the playing failure weight and the stalling weight corresponding to the non-business hot time section are set to be smaller.

[0139] Specifically, the playing failure weight and the stalling weight are both 10-100, the value of the playing failure weight and the stalling weight corresponding to the non-business hot time section can be in the range of 10-50, and the value of the playing failure weight and the stalling weight corresponding to the business hot time section can be in the range of 60-100.

[0140] It is worth noting that the value of the playing failure weight and the stalling weight is not fixed in the non-business hot time section or the business hot time section, but needs to be continuously increased with the serial number of the collection time section. For example, for the playing failure weight, for example, from 8 pm to 12 pm, the playing failure weight corresponding to 8 pm-9 pm can be 60, the playing failure weight corresponding to 9 pm-10 pm can be 70, and the playing failure weight corresponding to 10 pm-10 pm can be 80, and so on.

[0141] Deno is a scoring constant, the value of Deno needs to be determined based on the preset convergence interval, and the final quality evaluation score needs to be converged into the preset convergence interval.

[0142] For example, if the obtained current quality evaluation score is 449 and the convergence interval is 0-100, the specific value of the scoring constant can be determined based on the current quality evaluation score and the convergence interval at this time.

[0143] For example, assuming that there are 4 collection time segments, the corresponding stall weight, stall rate, play failure rate and play failure weight of each collection time segment are as follows:

[0144] Kdrate1=0.1,PlaydRate1=0.1,w1=0.25,m1=0.25

[0145] Kdrate2=0.2,PlaydRate2=0.2,w2=0.25,m2=0.25

[0146] Kdrate3=0.3,PlaydRate3=0.3,w3=0.25,m3=0.25

[0147] Kadate4=0.2,PlaydRate4=0.2,w4=0.25,m4=0.25

[0148] Then Q=(0.1*0.25+0.2*0.25+0.3*0.25+0.2*0.25+0.1*0.25+0.2*0.25+0.3*0.25+0.2*0.25) / 4*Deno

[0149] =(0.2+0.2) / 4*Deno

[0150] =0.1*(Deno=10000)

[0151] =1000

[0152] It is worth noting that when Deno is 10000, the evaluation score can be converged into the preset convergence interval.

[0153] In this embodiment, when calculating the quality evaluation score, the play failure weight corresponding to each collection time segment is different, and the stall weight corresponding to each collection time segment is also different, which fully considers the actual play situation, and thus ensures the play quality evaluation accuracy.

[0154] After determining the quality evaluation score, the evaluation unit 23 is configured to evaluate the play quality by using the quality evaluation score. Specifically, it includes:

[0155] determining a score interval range to which the quality evaluation score belongs;

[0156] finding a corresponding quality level from a mapping table according to the score interval range, the mapping table pre-storing a corresponding relationship between each score interval range and quality level.

[0157] Specifically, the score interval range includes: [1200, 1100), [1100, 900), [900, 700), [700, 400), [400, 0); the quality level corresponding to [1200, 1100) is excellent, the quality level corresponding to [1100, 900) is good, the quality level corresponding to [900, 700) is general, the quality level corresponding to [700, 400) is poor, and the quality level corresponding to [400, 0) is very poor.

[0158] Therefore, if it is determined that the interval range to which the quality evaluation score belongs is [1200, 1100), it is determined that the playing quality is excellent; if it is determined that the interval range to which the quality evaluation score belongs is [1100, 900), it is determined that the playing quality is good; if it is determined that the interval range to which the quality evaluation score belongs is [900, 700), it is determined that the playing quality is general; if it is determined that the interval range to which the quality evaluation score belongs is [700, 400), it is determined that the playing quality is poor; and if it is determined that the interval range to which the quality evaluation score belongs is [400, 0), it is determined that the playing quality is very poor.

[0159] As described in the foregoing example, when Q is 1000 and it is determined that the interval range to which the quality evaluation score Q belongs is [1100, 900), it is determined that the playing quality of the live broadcast platform is good.

[0160] The method for evaluating playing quality provided in this embodiment uses the collection data of the playing end as a data source to obtain stream media playing quality evaluation data (playing lag data and playing failure data), in which the proportion of user subjective feedback is very small and the proportion of objective parameters is large, so that the evaluation quality of the stream media playing can be ensured, and the quality evaluation precision is improved.

[0161] Based on the same inventive concept as the foregoing embodiments, this embodiment also provides a computer device 300, as shown in Figure 3 which includes a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320, and the processor 320 implements the following steps when executing the computer program 311:

[0162] obtaining playing failure data and playing lag data;

[0163] determining a playing failure rate based on the playing failure data;

[0164] determine a play failure rate based on the play failure data;

[0165] determine a quality evaluation score based on the play failure rate and the play hang rate;

[0166] evaluate the play quality by using the quality evaluation score.

[0167] In a specific implementation process, the processor 420 can implement any of the embodiments of the foregoing embodiments when executing the computer program 411.

[0168] Since the computer device introduced in the embodiment is a device used to implement the method for evaluating the play quality, based on the method introduced in the first embodiment of the present application, those skilled in the art can understand the specific implementation of the computer device of the present embodiment and its various forms, so the server how to implement the method in the embodiment of the present application will not be introduced in detail. As long as the device used to implement the method in the embodiment of the present application is implemented by those skilled in the art, it belongs to the scope of the present application.

[0169] Based on the same inventive concept, the present application also provides a computer readable storage medium 400, as shown in the figure, which stores a computer program 411, and the computer program 411 is executed by a processor to implement the following steps: Figure 4

[0170] obtain play failure data and play hang data;

[0171] determine a play failure rate based on the play failure data;

[0172] determine a play hang rate based on the play hang data;

[0173] determine a quality evaluation score based on the play failure rate and the play hang rate;

[0174] evaluate the play quality by using the quality evaluation score.

[0175] In a specific implementation process, the computer program 411 can implement any of the embodiments of the foregoing embodiments when executed by the processor.

[0176] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.​

[0177] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0178] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0179] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0180] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the embodiments by those of skill in the art once they have the benefit of the present disclosure. Therefore, the appended claims are intended to cover all such variations and modifications that fall within the scope of the application.

[0181] The method, device, medium and computer device for evaluating playing quality provided by the present application can bring at least the following beneficial effects:

[0182] The application discloses a method, device, medium and computer equipment for evaluating playing quality. The method comprises the following steps: acquiring playing failure data and playing lag data; determining a playing failure rate based on the playing failure data; determining a playing lag rate based on the playing lag data; determining a quality evaluation score based on the playing failure rate and the playing lag rate; and evaluating playing quality by using the quality evaluation score. Thus, the application uses the collected data of a playing end as a data source to obtain streaming media playing quality evaluation data (playing lag data and playing failure data). The subjective feedback of users accounts for a small proportion in the data, and the objective parameters account for a large proportion. Therefore, the evaluation quality of the streaming media playing can be ensured, and the quality evaluation precision is improved. Moreover, the method provided by the embodiment does not need to depend on an actual scene, has strong universality, and has high popularization value.

[0183] The above merely describes preferred embodiments of the application, but should not be used to limit the protection scope of the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application should be included in the protection scope of the application.

Claims

1. A method for evaluating a quality of play, characterized by, The method comprises: acquiring playing failure data and playing lag data; determining a playing failure rate based on the playing failure data; determining a playing lag rate based on the playing lag data; determining a quality evaluation score based on the playing failure rate and the playing lag rate; evaluating playing quality by using the quality evaluation score; the acquiring playing failure data comprises: acquiring the number of successfully played videos, the number of failed videos and the first total amount of video acquisition within a preset acquisition time segment; the acquiring playing lag data comprises: acquiring the number of videos with playing lag, the lag duration and the second total amount of video acquisition within a preset acquisition time segment; the determining a playing failure rate based on the playing failure data comprises: According to the formula determining the play failure rate Playrate; wherein the Sf i is the number of videos that fail to play in the i-th collection time segment; the St i is the corresponding first video collection total in the i-th collection time segment; the N is the total number of collection time segments, and the i is the serial number of each collection time segment; the determining a playing lag rate based on the playing lag data comprises: According to the formula determining the playback stall rate Kdrate; wherein Kd i is the number of videos with playback stalls in the ith collection time segment; the Ht i is the second video collection total corresponding to the ith collection time segment, the Kdt i is the stall duration corresponding to the ith collection time segment, the Time is the collection duration corresponding to the ith collection time segment, the N is the total number of collection time segments, and the i is the serial number of each collection time segment.

2. The method of claim 1, wherein, the determining a quality evaluation score based on the playing failure rate and the playing lag rate comprises: According to the formula determining the quality evaluation score Q; wherein, the w i Kdrate is the stalling weight corresponding to the i-th collection time segment, Kdrate i is the stalling rate corresponding to the i-th collection time segment, the m i Playdrate is the playing failure weight corresponding to the i-th collection time segment, Playdrate i is the playing failure rate corresponding to the i-th collection time segment, the Deno is a scoring constant, the N is the total number of collection time segments, and the i is the serial number of each collection time segment.

3. The method of claim 1, wherein, the evaluating playing quality by using the quality evaluation score comprises: determining a score interval range to which the quality evaluation score belongs; looking up a corresponding quality level from a mapping table according to the score interval range; the mapping table has pre-stored corresponding relationships between each score interval range and the quality level.

4. A storage medium having stored thereon a computer program, characterized in that The program is executed by a processor to implement the method of any one of claims 1 to 3.

5. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the method of any one of claims 1 to 3.

Images