Method, system and equipment for processing poor user quality of mobile network video and medium

Abstract

The invention provides a mobile network video user poor-quality processing method and system, electronic equipment and a storage medium, and aims to solve the problem that poor-quality user identification is poor in flexibility and accuracy, and the method comprises the steps: obtaining original user detailed list data of a video user; performing data preprocessing on the detailed list data; obtaining experience index data of each video user through the preprocessed detailed list data so as to carry out threshold analysis on video user experience indexes, and establishing a video user experience index score table; analyzing the video user service duration data through the preprocessed original user detailed list data, and establishing a video user service duration score table; calculating an acceleration guarantee recommendation value according to the video user experience index score table and the video user service duration score table, and obtaining an acceleration guarantee recommendation value table of the video user; and selecting a video user according to the acceleration guarantee recommendation value table to perform network acceleration. According to the invention, the accuracy and evaluation adaptability of poor-quality user identification can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of network communication technologies, and particularly to a method for processing poor-quality mobile network video users, a system for processing poor-quality mobile network video users, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the rapid development of mobile Internet and streaming media services, users' requirements for the quality of video services on mobile networks are increasing day by day. How to effectively evaluate and optimize the quality of experience of video users has become an important topic in the field of mobile communication. The quality of experience of mobile network video services is usually quantified by a series of key quality indicators (KQIs), such as download rate, latency, and stutter frequency.

[0003] Most traditional methods use static thresholds to judge the quality of experience. However, due to the complexity and instability of the mobile network environment, the static threshold method cannot adapt to changes in different network conditions, resulting in poor flexibility and accuracy in identifying poor-quality users. Secondly, existing technologies often lack specialized analysis of the key quality indicators of video users and are difficult to accurately reflect the specific experience of video viewing. In addition, most existing methods do not consider the video service behavior data of users, making it difficult for experience evaluation to comprehensively reflect users' preferences and needs for video services. Finally, existing technologies usually only identify poor-quality users and lack effective optimization and resource allocation means, and cannot provide necessary network acceleration and guarantee measures for users with poor experience in a timely manner. Summary of the Invention

[0004] In order to at least solve the problem in the prior art that a static threshold is set to identify users with poor experience, which cannot adapt to changes in different network conditions, resulting in poor flexibility and accuracy in identifying poor-quality users. The present disclosure provides a method for processing poor-quality mobile network video users, a system for processing poor-quality mobile network video users, an electronic device, and a computer-readable storage medium, which can more comprehensively analyze and locate users with poor experience, enhance the accuracy and rationality of scoring, and avoid the one-sidedness of simply relying on network indicators. The system can flexibly respond to changes in different network conditions, avoid evaluation errors under static thresholds, and improve the accuracy of identifying poor-quality users and the evaluation adaptability.

[0005] In a first aspect, the present disclosure provides a method for processing poor-quality mobile network video users, the method comprising:

[0006] Obtaining the original user detail data of cellular mobile network accelerated video users;

[0007] Performing data preprocessing on the original user detail data;

[0008] Obtaining experience index data of each video user through the pre-processed original user detail data, so as to perform threshold analysis of the video user experience index and establish a video user experience index scoring table; and,

[0009] Analyze the service duration data of video users through the pre-processed original user detailed data, and establish a video user service duration score table;

[0010] Calculate the acceleration guarantee recommendation value based on the video user experience index scoring table and the video user service duration scoring table, and obtain the acceleration guarantee recommendation value table for video users;

[0011] Select video users for network acceleration based on the acceleration guarantee recommended value table.

[0012] Furthermore, the original user detailed order data includes:

[0013] User-related fields, protocol size class, APPID, effective download rate, XKB (X Keyboard Extension) startup delay, traffic and time.

[0014] Furthermore, the data preprocessing of the original user detail list data includes:

[0015] Extract video user details: match the video user list data in the cellular mobile network according to user-related fields, protocol size class and APPID;

[0016] Data cleaning: Detect video user list data and remove outliers in the data, and use mean filling to supplement the default values ​​in the data;

[0017] Granularity selection and acquisition of video user service duration data: Select a preset time length as the time granularity, count the viewing service duration of each video user in the video user list data after data cleaning, obtain video user service duration data and use it as a behavioral indicator to evaluate video user experience.

[0018] Furthermore,

[0019] The experience indicators include effective download rate and XKB startup delay;

[0020] The pre-processed original user detailed data is used to obtain the experience index data of each video user, so as to perform threshold analysis of the video user experience index and establish a video user experience index scoring table, including:

[0021] Analyze the effective download rate of video users and the distribution of XKB startup latency through the preprocessed original user detail data, determine the thresholds of the effective download rate and XKB startup latency, and divide the effective download rate and XKB startup latency into multiple scoring intervals respectively;

[0022] Determine the low-rate recommended value interval and the high-latency recommended value interval, and determine the scoring values corresponding to each interval of the effective download rate and each interval of XKB startup latency respectively;

[0023] Based on the scoring intervals and scoring value ranges of the effective download rate and XKB startup latency, formulate a scoring table for video user experience indicators.

[0024] Furthermore, analyze the video user service duration data through the preprocessed original user detail data, and establish a scoring table for video user service duration, including:

[0025] According to actual business experience, divide the service duration of video users into multiple scoring intervals through the preprocessed original user detail data to reflect the activity and continuity of video users watching videos, and set corresponding scoring values for each scoring interval;

[0026] Based on the scoring intervals and scoring values of the service duration, formulate a scoring table for video user service duration.

[0027] Furthermore, calculate the acceleration guarantee recommended value according to the video user experience indicator scoring table and the video user service duration scoring table, and obtain the acceleration guarantee recommended value table for video users, including:

[0028] S1: Preset the weights of the effective download rate and XKB startup latency to a and b respectively, and the weights of the video user experience indicator and video user service duration to c and d respectively, and a + b = 1, c + d = 1;

[0029] S2: According to the obtained video user experience indicator scoring table, calculate the experience value KQI, and the specific formula is as follows:

[0030] KQI = LR × a + HL × b;

[0031] Among them, LR represents the scoring value corresponding to the interval where the effective download rate of video users is located, and HL represents the scoring value corresponding to the interval where the XKB startup latency of video users is located;

[0032] S3: According to the obtained video user service duration scoring table, obtain the service duration scoring value, and combine the experience value KQI to calculate the acceleration guarantee recommended value RV, and the specific formula is as follows:

[0033] RV = KQI × c + T × d;

[0034] Among them, T is the scoring value corresponding to the interval where the video user service duration is located;

[0035] S4: Obtain the video user acceleration guarantee recommendation value table according to the acceleration guarantee recommendation value RV.

[0036] Furthermore, the selecting video users for network acceleration according to the acceleration guarantee recommendation value table includes:

[0037] Perform video user slice statistics verification according to the acceleration guarantee recommendation value table to obtain the acceleration guarantee recommendation value threshold, count the acceleration guarantee recommendation times of video users according to the acceleration guarantee recommendation value threshold, and select video users to implement acceleration according to the acceleration guarantee recommendation times of video users.

[0038] Furthermore, the counting the acceleration guarantee recommendation times of video users according to the acceleration guarantee recommendation value threshold and selecting video users to implement acceleration according to the acceleration guarantee recommendation times of video users includes:

[0039] Taking the preset time length as the granularity, count the number of times that the acceleration guarantee recommendation value of video users is greater than or equal to the acceleration guarantee recommendation value threshold within a certain time to obtain the acceleration guarantee recommendation times;

[0040] Sort the acceleration guarantee recommendation times of video users within a certain time from high to low, and select users whose acceleration guarantee recommendation times are in the top certain proportion of the total number of video users to implement acceleration, so as to ensure that rate guarantee is provided preferentially for users with poor experience.

[0041] In a second aspect, the present disclosure provides a system for processing poor quality of mobile network video users, and the system includes:

[0042] An acquisition module, which is set to acquire the original user detail data of video users accelerated by the cellular mobile network;

[0043] A preprocessing module, which is set to perform data preprocessing on the original user detail data;

[0044] An experience index scoring module, which is set to obtain the experience index data of each video user through the preprocessed original user detail data, perform threshold analysis on the video user experience index, and establish a video user experience index scoring table; and,

[0045] A service duration scoring module, which is set to analyze the video user service duration data through the preprocessed original user detail data and establish a video user service duration scoring table;

[0046] A recommendation value determination module, which is set to calculate the acceleration guarantee recommendation value according to the video user experience index scoring table and the video user service duration scoring table, and obtain the acceleration guarantee recommendation value table of video users;

[0047] An acceleration module, which is configured to select video users for network acceleration according to an acceleration guarantee recommendation value table.

[0048] In a third aspect, the present disclosure provides an electronic device, including a memory and a processor. A computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes the mobile network video user quality degradation processing method as described in any one of the first aspects.

[0049] In a fourth aspect, the present disclosure provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the mobile network video user quality degradation processing method as described in any one of the first aspects is implemented.

[0050] Beneficial effects:

[0051] The mobile network video user quality degradation processing method, mobile network video user quality degradation processing system, electronic device and storage medium provided by the present disclosure can adjust the KQI threshold according to the actual network environment and user distribution, ensuring the accuracy and flexibility of identifying quality degradation users. At the same time, by combining key KQI indicators of video users and behavioral data such as service duration, it can more objectively and comprehensively reflect the video service experience of video users. By screening users for priority acceleration through the acceleration guarantee recommendation value, effective allocation of resources is achieved, ensuring that users with poor experience can obtain timely service guarantee under dynamically changing network conditions, and improving the overall user experience and network resource utilization rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a flowchart of a mobile network video user quality degradation processing method provided in Embodiment 1 of the present disclosure;

[0053] Figure 2 It is a flowchart of a mobile network video user quality degradation processing method provided in Embodiment 2 of the present disclosure;

[0054] Figure 3 It is an architecture diagram of a mobile network video user quality degradation processing system provided in Embodiment 3 of the present disclosure;

[0055] Figure 4 It is an architecture diagram of an electronic device provided in Embodiment 4 of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] To enable those skilled in the art to better understand the technical solutions of the present disclosure, the present disclosure will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments and accompanying drawings described herein are only for explaining the present invention and are not intended to limit the present invention.

[0057] It should be noted that the terms "first", "second", etc. in the description, claims and above-mentioned drawings of the present disclosure are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence; moreover, without conflict, the embodiments in the present disclosure and the features in the embodiments can be arbitrarily combined with each other.

[0058] Among them, the terms used in the embodiments of the present disclosure are only for the purpose of describing specific embodiments, and are not intended to limit the present disclosure. The singular forms "a", "the" and "said" used in the embodiments of the present disclosure and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0059] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only for the convenience of the description of the present disclosure, and have no specific meaning in themselves. Therefore, "module", "component" or "unit" can be used interchangeably.

[0060] The following uses specific embodiments to detail the technical solutions of the present disclosure and how the technical solutions of the present disclosure solve the technical problems existing in the prior art. It can be understood that in the embodiments of the present application, the execution subject can execute some or all of the steps in the embodiments of the present application. These steps or operations are only examples, and the embodiments of the present application can also execute other operations or various deformations of the operations. In addition, the various steps can be executed in different orders presented in the embodiments of the present application, and it is possible not to execute all the operations in the embodiments of the present application. Moreover, the following several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0061] Figure 1 It is a schematic flowchart of a method for processing poor quality of mobile network video users provided in Embodiment 1 of the present disclosure, as Figure 1 shown, the method includes:

[0062] Step S101: Obtain the original user detail data of video users with cellular mobile network acceleration;

[0063] Step S102: Perform data preprocessing on the original user detail data;

[0064] Step S103: Through the preprocessed original user detail data, obtain the experience index data of each video user to perform threshold analysis on the video user experience index, and establish a video user experience index scoring table; and,

[0065] Step S104: Through the preprocessed original user detail data, analyze the video user service duration data and establish a video user service duration scoring table;

[0066] Step S105: Calculate the acceleration guarantee recommendation value according to the video user experience index scoring table and the video user service duration scoring table, and obtain the acceleration guarantee recommendation value table for video users;

[0067] Step S106: Select video users for network acceleration according to the acceleration guarantee recommendation value table.

[0068] The purpose of the embodiments of the present disclosure is to effectively identify video service experience poor users in the cellular mobile network acceleration user group, quantify their acceleration requirement degree, intelligently recommend corresponding acceleration strategies, and improve the experience of video users (referred to as users for short). Due to the limitation of network resources, in order to achieve the best network acceleration effect, it is necessary to conduct poor quality analysis on mobile network video users, so as to fairly and accurately screen out the video users who most need to be accelerated and improved first, and perform network acceleration processing to improve the network resource utilization rate as much as possible.

[0069] In the present disclosure, the cellular mobile network acceleration refers to the behavior of watching videos through the underlying cellular mobile network (non-Wi-Fi conditions such as using 4G, 5G and other cellular mobile networks). Video users are all users who need to be accelerated through the cellular mobile network. The cellular mobile network acceleration improves the user's video watching experience and the like through the cellular mobile network performance.

[0070] To achieve the purpose of the embodiments of the present disclosure, this embodiment first obtains the original user detail list data of video users through data collection, that is, all data related to the user's video watching, downloading and other behaviors; for the original user detail list data, preprocessing is required to remove abnormal data therein, improve data quality, unify data formats, etc.

[0071] Through the preprocessed original user detail list data, obtain the experience index data of each video user. The user's experience index is an index that can reflect the video loading time, caching time, smoothness and response speed during video playback, etc. For the obtained experience index data, explore the threshold of the influence of video watching according to the user distribution and the actual network condition corresponding to the experience index data, determine the interval of the low rate recommendation value and the interval of the high delay recommendation value, and obtain the video user experience index scoring table; thus, through dynamic threshold exploration, adjust the threshold value of the experience index KQI according to the changes of user distribution and network conditions. Compared with the traditional fixed threshold method, the dynamic threshold can adaptively cope with different network fluctuations and user requirements.

[0072] Through user order data, we can also obtain the service time data of users watching videos, analyze the service time data of video users, and establish a video user service time rating table; by introducing the service time behavior data, we can more comprehensively and accurately evaluate the actual video service needs of video users, and reflect the importance of users' demand for mobile network acceleration.

[0073] Then, the acceleration guarantee recommendation value is calculated based on the video user experience index score table and the video user service duration score table, and the acceleration guarantee recommendation value table for video users is obtained. Video users are selected for network acceleration based on the acceleration guarantee recommendation value table. The acceleration guarantee recommendation value table takes into account both the demand for service experience improvement and the video user's preference for video services, so as to more comprehensively and accurately evaluate the actual video service needs of video users, thereby screening out the video users who need priority acceleration improvement.

[0074] The disclosed embodiment adjusts the KQI threshold according to the actual network environment and user distribution to ensure the accuracy and flexibility of poor quality user identification. At the same time, combined with the key KQI indicators of video users and behavioral data such as service duration, it more objectively and comprehensively reflects the video service experience of video users. By selecting priority users for acceleration through the acceleration guarantee recommendation value, effective resource allocation is achieved, ensuring that users with poor experience receive timely service guarantees under dynamically changing network conditions, thereby improving the overall user experience and network resource utilization.

[0075] Furthermore, the original user detailed order data includes:

[0076] User-related fields, protocol size class, APPID, effective download rate, XKB startup delay, traffic and time.

[0077] Through log collection, API (Application Programming Interface), or related database query, all data related to the user's viewing of videos can be obtained, which is convenient for subsequent user data statistical analysis.

[0078] Furthermore, the data preprocessing of the original user detail list data includes:

[0079] Extract video user details: match the video user list data in the cellular mobile network according to user-related fields, protocol size class and APPID;

[0080] Data cleaning: Detect video user list data and remove outliers in the data, and use mean filling to supplement the default values ​​in the data;

[0081] Granularity Selection and Video User Service Duration Data Acquisition: Select a preset time length as the time granularity, and count the viewing service duration of each video user from the video user list data after data cleaning to obtain the video user service duration data, which is used as a behavioral indicator for evaluating video user experience.

[0082] Extract the video user detail list and match each user. Obtain all video viewing-related data of the user through user-related fields, protocol size categories, and APP ID, which is convenient for subsequent statistical analysis. Ensure the rationality and integrity of the data distribution through data cleaning; conduct statistical analysis on the video user list data after data cleaning. Select a preset time length as the time granularity. The preset time length can be selected according to the actual situation, such as one hour, two hours, or three hours, etc. For the convenience of statistics, two hours is preferably used as the time granularity, which can effectively balance the requirements of real-time monitoring and data stability, timely reflect the dynamic changes of user experience, and make the network quality guarantee more targeted and practical. Taking two hours as the time granularity, count the viewing service duration of each cellular mobile network video user as a behavioral indicator for evaluating video user experience, and obtain the video user service duration data.

[0083] Furthermore,

[0084] The experience indicators include effective download rate and XKB startup latency;

[0085] From the preprocessed original user detail list data, obtain the experience indicator data of each video user to conduct threshold analysis of video user experience indicators, and establish a video user experience indicator scoring table, including:

[0086] Analyze the distribution of the effective download rate and XKB startup latency of video users through the preprocessed original user detail list data, determine the thresholds of the effective download rate and XKB startup latency, and divide the effective download rate and XKB startup latency into multiple scoring intervals respectively;

[0087] Determine the scoring values corresponding to each interval of the effective download rate and each interval of the XKB startup latency respectively;

[0088] Based on the scoring intervals and scoring value ranges of the effective download rate and XKB startup latency, formulate a video user experience indicator scoring table.

[0089] Set two key KQI indicators, the effective download rate and XKB startup latency, which can directly reflect the viewing experience of video users; the effective download rate is used to measure the smoothness of the video playback process, and the startup latency reflects the response speed of the video startup after the user clicks to play. Of course, other experience indicators can also be added, such as buffering time, video loading time, frame drop rate, playback smoothness, packet loss rate, etc., which can be set according to the actual situation.

[0090] Based on the video user inventory data, obtain the effective download rate and the XKB startup latency, conduct threshold exploration for video user experience metrics, and establish a scoring table for video user experience metrics. The specific steps are as follows:

[0091] According to the distribution of the user experience metrics, determine the reasonable thresholds for the effective download rate and the XKB startup latency. Divide the effective download rate into multiple scoring intervals, and also divide the XKB startup latency into multiple scoring intervals, such as 4 - 10 intervals. Preferably, set 7 scoring intervals for the effective download rate and 5 scoring intervals for the XKB startup latency. Analyze which intervals may lead to poor user experience according to the actual situation, and respectively determine the low - rate recommended value intervals and the high - latency recommended value intervals to ensure the dynamic adaptability of the thresholds, and conduct score setting for each interval. Higher scores will be set for the low - rate recommended value intervals and the high - latency recommended value intervals. Thus, conduct interval division and assignment according to the user distribution and the actual experience of each corresponding value. An exemplary score setting is that the corresponding score value range for the effective download rate is [0, 10]. The lower the effective download rate, the higher the score value, and vice versa. The corresponding score value range for the XKB startup latency is [0, 10]. The higher the XKB startup latency, the higher the score value, and vice versa; the score values are all taken as integers.

[0092] Based on the exploration results of the threshold values, formulate a scoring table for video user experience metrics. The scoring table is used to convert the effective download rate and the startup latency into specific scores to quantify the video user experience. An exemplary video user experience metrics scoring table is shown in Table 1 below.

[0093] Table 1: Scoring Table for Video User Experience Metrics

[0094]

[0095] Through dynamic threshold exploration, combined with the user distribution, determine the reasonable threshold values for each KQI metric, so that the system can dynamically adjust the evaluation criteria according to the network environment and user distribution. Compared with the traditional fixed - threshold scheme, the dynamic threshold is more flexible and can cope with problems such as different line networks, regions, times, and user requirements, and adapt to the fluctuating changes of mobile network conditions and user experience.

[0096] Furthermore, analyze the video user service duration data through the pre - processed original user detail list data, and establish a scoring table for video user service duration, including:

[0097] According to actual business experience, divide the service duration of video users into multiple scoring intervals through the pre - processed original user detail list data to reflect the activity and continuity of video users watching videos, and set corresponding score values for each scoring interval;

[0098] Based on the scoring intervals and scores of business duration, formulate a scoring table for video user business duration.

[0099] In order to obtain the business demand status of users for watching videos, according to actual business experience, the business duration is divided into multiple scoring intervals, such as 3, and scored to reflect the activity and continuity of video users watching videos. For example, for the two-hour granularity, that is, the business duration of 120 minutes is analyzed. When the business duration is in the range of (60, 120], the score value is 10; when the business duration is in the range of (20, 60], the score value is 8; when the business duration is in the range of (6, 20], the score value is 4; when the business duration is in the range of [0, 6], it is not included in the analysis scope.

[0100] Based on the exploration results of the threshold value, formulate a scoring table for video user business duration, as shown in Table 2 below, to quantify the behavior of video users.

[0101] Table 2: Example of Scoring Table for Video User Business Duration

[0102]

[0103] Furthermore, calculate the accelerated guarantee recommendation value according to the video user experience index scoring table and the video user business duration scoring table, and obtain the accelerated guarantee recommendation value table for video users, including:

[0104] S1: Preset the weights of the effective download rate and the XKB startup delay to be a and b respectively, and the weights of the video user experience index and the video user business duration to be c and d respectively, and a + b = 1, c + d = 1;

[0105] S2: According to the obtained video user experience index scoring table, calculate the experience value KQI. The specific formula is as follows:

[0106] KQI = LR × a + HL × b;

[0107] Among them, LR represents the score value corresponding to the interval where the video user's effective download rate is located, and HL represents the score value corresponding to the interval where the video user's XKB startup delay is located;

[0108] S3: According to the obtained video user business duration scoring table, obtain the business duration score value, and combine it with the experience value KQI to calculate the accelerated guarantee recommendation value RV. The specific formula is as follows:

[0109] RV = KQI × c + T × d;

[0110] Among them, T is the score value corresponding to the interval where the video user's business duration is located;

[0111] S4: Obtain the accelerated guarantee recommendation value table for video users according to the accelerated guarantee recommendation value RV.

[0112] Each weight value can be set based on actual experience. For example, the weight a of the effective download rate is set to 0.5 - 0.8; the weight b of the XKB startup delay is set to 0.2 - 0.5, preferably a is set to 0.6 and b is 0.4. Similarly, the weight c of the user experience index can be set to 0.6 - 0.9, preferably 0.7, and the weight d of the video user service duration can be set to 0.1 - 0.4, preferably 0.3.

[0113] According to the formula, a table containing the recommended values for accelerating the protection of each video user can be calculated. Among them, for the experience value, the experience value is excellent when it is between (9, 10], good when it is between (8, 9], medium when it is between (6, 8], and poor when it is between [0, 6]; for the business duration score value, it is long when it is between (8, 10], medium when it is between (4, 8], and short when it is between (2, 4].

[0114] By assigning weights to the experience index and the business duration during the scoring process of video users, the weighted scoring mechanism can balance the influence of each index, avoid the deviation of a single index from affecting the overall score, thereby ensuring the fairness and accuracy of the recognition result, and enabling the score to comprehensively and objectively reflect the viewing experience of video users.

[0115] Furthermore, the selection of video users for network acceleration according to the recommended value table for accelerating protection includes:

[0116] Perform slice statistics verification on video users according to the recommended value table for accelerating protection, obtain the threshold of the recommended value for accelerating protection, count the number of recommended times for accelerating protection of video users according to the threshold of the recommended value for accelerating protection, and select video users for acceleration according to the number of recommended times for accelerating protection of video users.

[0117] Through the obtained recommended value table for accelerating protection, perform slice statistics on video users with different scores to obtain the threshold of the recommended value for accelerating protection. Since network resources are limited, by counting the number of users in each score interval of the recommended value for accelerating protection, determine the users who can be accelerated, and determine the threshold of the recommended value for guarantee protection. Taking the above preferred values as an example, through actual verification, when the threshold of the recommended value for accelerating protection is set to 3, the overall accelerating protection effect is optimal. By counting the number of times exceeding the threshold of the recommended value for guarantee protection within a certain period of time, it indicates that during this period, the user has a greater demand for network acceleration, so as to screen out the users who need to be accelerated and improved first, ensure that the acceleration resources are allocated to the users with the worst experience, improve the utilization rate of network resources, and regularly adjust the acceleration strategy according to network conditions.

[0118] Furthermore, the counting of the number of recommended times for accelerating protection of video users according to the threshold of the recommended value for accelerating protection, and the selection of video users for acceleration according to the number of recommended times for accelerating protection of video users includes:

[0119] Taking the preset time length as the granularity, count the number of times that the acceleration guarantee recommendation value of a video user is greater than or equal to the acceleration guarantee recommendation value threshold within a certain period of time, and obtain the acceleration guarantee recommendation count;

[0120] Sort the acceleration guarantee recommendation counts of video users within a certain period of time from high to low, and select users whose acceleration guarantee recommendation counts are in a certain proportion of the total number of video users to implement acceleration, so as to ensure that bitrate guarantee is provided preferentially for users with a worse experience.

[0121] Taking the preset time length as the granularity, such as 2 hours, count the number of times that the acceleration guarantee recommendation value of a video user is greater than or equal to the acceleration guarantee recommendation value threshold 3 within a certain period of time (such as within each week or within half a month), and obtain the acceleration guarantee recommendation count. Taking one week as an example, sort the acceleration guarantee recommendation counts of video users from high to low every week, and select users whose acceleration guarantee recommendation counts are in a certain proportion (selected according to resource conditions, such as 30%, 40%, 50% or 60%) of the total number of users with acceleration guarantee in the cellular mobile network to implement acceleration, ensuring that bitrate guarantee is provided preferentially for users with a worse experience. Through methods such as adaptive bitrate, bandwidth prediction, and priority scheduling, ensure that the video stream can be transmitted at a stable bitrate under different network conditions, so as to provide a high-quality video experience

[0122] According to the acceleration guarantee recommendation value, screen out the video users who most need to be accelerated and improved preferentially, which can ensure that the acceleration resources are allocated to the users with the worst experience, further improve the utilization rate of network resources, and enable the limited network and service resources to more effectively improve the user experience. At the same time, the regular calculation and statistics of the acceleration recommendation value can adapt to the dynamic changes of the network and user behaviors, ensuring the timeliness and accuracy of the acceleration strategy.

[0123] The embodiments of the present disclosure can more comprehensively analyze and locate users with poor experience by comprehensively considering the behavior data of users, enhance the accuracy and rationality of the scoring, and avoid the one-sidedness of simply relying on network metrics. At the same time, the dynamic threshold mechanism can enable the system to flexibly respond to different network condition changes, avoid the evaluation error under the static threshold, and improve the accuracy and evaluation adaptability of identifying users with poor quality. By calculating the acceleration guarantee recommendation value, screen out the users who need to be accelerated and improved preferentially, ensure that the acceleration resources are allocated to the users with the worst experience, and regularly adjust the acceleration strategy according to the network conditions, improving the utilization rate of network resources.

[0124] Embodiment 2 of the present disclosure also provides a method for processing poor quality of mobile network video users, such as Figure 2 shown, the method includes the following steps:

[0125] 1) Data collection:

[0126] Obtain the original user detail data of cellular mobile network acceleration users (i.e. video users), including user-related fields, protocol size class, APPID, effective download rate, XKB startup delay, traffic and time.

[0127] 2) Preprocessing the original user details of the cellular mobile network acceleration user:

[0128] Data preprocessing is performed on the original user list data of cellular mobile network acceleration users, including matching the inherited video user list data, data cleaning and statistical granularity selection, to obtain the video user service duration data, as follows:

[0129] a. Extract video user details: Match the video user list data in the cellular mobile network according to user-related fields, protocol size class and APPID.

[0130] b. Data cleaning: Detect the video user list data and remove outliers in the data to ensure the rationality of data distribution. For default values, use mean filling to supplement them to ensure data integrity.

[0131] c. Granularity selection and video user service duration data acquisition: Perform statistical analysis on the video user list data after data cleaning. Selecting two hours as the time granularity can effectively balance the needs of real-time monitoring and data stability, timely reflect the dynamic changes in user experience, and make network quality assurance more targeted and practical. With two hours as the time granularity, count the viewing service time of each cellular mobile network video user as a behavioral indicator for evaluating video user experience, and obtain video user service duration data.

[0132] 3) Perform threshold exploration on the effective download rate and XKB startup delay, determine the interval of low rate recommended values ​​and the interval of high delay recommended values, and obtain the video user experience index score table. Analyze the video user service duration data and establish the video user service duration score table. The specific steps for obtaining the video user experience index score table and the video user service duration score table are as follows:

[0133] 3.1) Obtain the effective download rate and XKB startup delay through the video user list data in step 2), conduct threshold exploration of video user experience indicators, and establish a video user experience indicator scoring table. The specific steps are as follows:

[0134] 3.1.1) Determine reasonable thresholds for the effective download rate and the XKB startup latency according to the user distribution. Divide the effective download rate into 7 scoring intervals and the XKB startup latency into 5 scoring intervals. Calculate the low-rate recommended value interval and the high-latency recommended value interval respectively to ensure the dynamic adaptability of the thresholds. The corresponding scoring value range for the effective download rate is [0, 10]. The lower the effective download rate, the higher the scoring value, and vice versa. The corresponding scoring value range for the XKB startup latency is [0, 10]. The higher the XKB startup latency, the higher the scoring value, and vice versa. The scoring values are all integers.

[0135] 3.1.2) Based on the exploration results of the threshold values, formulate a scoring table for video user experience indicators, as shown in Table 1. The scoring table is used to convert the effective download rate and the startup latency into specific scores to quantify the experience of video users.

[0136] Table 1: Example of Scoring Table for Video User Experience Indicators

[0137]

[0138] 3.2) Through the video user service duration data in step 2), conduct threshold exploration for video user behavior indicators and establish a scoring table for video user service duration. The specific steps are as follows:

[0139] 3.2.1) According to actual business experience, divide the service duration into 3 scoring intervals to reflect the activity and continuity of video users watching videos. Analyze the service duration at a two-hour granularity, that is, 120 minutes. For the service duration in the range of (60, 120], the scoring value is 10; for the service duration in the range of (20, 60], the scoring value is 8; for the service duration in the range of (6, 20], the scoring value is 4; the service duration in the range of [0, 6] is not included in the analysis scope.

[0140] 3.2.2) Based on the exploration results of the threshold values, formulate a scoring table for video user service duration to quantify the behavior of video users.

[0141] Table 2: Example of Scoring Table for Video User Service Duration

[0142]

[0143] 4) Calculate the experience value and the acceleration guarantee recommended value according to the video user experience indicator scoring table and the video user service duration scoring table obtained in step 3), and obtain the acceleration guarantee recommended value table for cellular mobile network accelerated users. The specific steps are as follows:

[0144] 4.1) Based on actual business experience, the preset weights of the effective download rate and the XKB startup delay are 0.6 and 0.4 respectively, and the weights of the video user experience index and the video user service duration are 0.7 and 0.3 respectively.

[0145] 4.2) According to the video user experience index scoring table, calculate the experience value KQI. The specific formula is as follows:

[0146] S2: According to the obtained video user experience index scoring table, calculate the experience value KQI. The specific formula is as follows:

[0147] KQI = LR × 0.6 + HL × 0.4;

[0148] Among them, LR represents the scoring value corresponding to the interval where the effective download rate of the video user is located, and HL represents the scoring value corresponding to the interval where the XKB startup delay of the video user is located;

[0149] 4.3) According to the video user service duration scoring table, obtain the service duration scoring value. Combine it with the experience value KQI to calculate the acceleration guarantee recommendation value RV. The specific formula is as follows:

[0150] RV = KQI × 0.7 + T × 0.3;

[0151] Among them, T is the scoring value corresponding to the interval where the video user service duration is located;

[0152] 4.4) Obtain the acceleration guarantee recommendation value table for cellular mobile network acceleration users. Among them, for the experience value, the experience value in the range of (9, 10] is excellent, (8, 9] is good, (6, 8] is medium, and [0, 6] is poor; for the service duration scoring value, (8, 10] is long, (4, 8] is medium, and (2, 4] is short. The acceleration guarantee recommendation table is shown in Table 3.

[0153] Table 3: Acceleration guarantee recommendation value table for cellular mobile network acceleration users

[0154] Experience value Business duration score value Acceleration guarantee recommendation value Poor Long 10 Poor Medium 9 Poor Short 8 Medium Long 7 Medium Medium 6 Medium Short 5 Good Long 4 Good Medium 3 Good Short 2 Excellent Long 1 Excellent Medium 0 Excellent Short 0

[0155] 5) Based on the acceleration guarantee recommendation value obtained in step 4) and the acceleration guarantee recommendation value table for cellular mobile network acceleration users, conduct video user slice statistics verification, obtain the acceleration guarantee recommendation value threshold, count the acceleration guarantee recommendation times, select users with high times for acceleration, so as to complete the quality difference analysis and guarantee of mobile network video users. The specific steps are as follows:

[0156] a2: Based on the acceleration guarantee recommendation value obtained in step S4 and the cellular mobile network acceleration guarantee recommendation value table, conduct slice statistics on video users with different scores to obtain the acceleration guarantee recommendation value threshold. After actual verification, when the acceleration guarantee recommendation value threshold is set to 3, the overall acceleration guarantee effect is the best.

[0157] b2: Count the number of times the acceleration guarantee recommendation value is greater than or equal to 3 for each video user within each week at a two-hour granularity to obtain the acceleration guarantee recommendation count.

[0158] c3: Sort the acceleration guarantee recommendation counts of the cellular mobile network acceleration guarantee users from high to low every week, and select the users whose acceleration guarantee recommendation counts are in the top 50% of the total number of cellular mobile network acceleration guarantee users to implement acceleration, ensuring that bitrate guarantee is provided preferentially for users with a poor experience.

[0159] The embodiments of the present disclosure are directed to video users among the cellular mobile network acceleration users, and combine two key KQI indicators, namely the effective download rate and the XKB startup delay, which directly reflect the viewing experience of video users. The effective download rate is used to measure the smoothness of the video playback process, and the startup delay reflects the response speed of the video startup after the user clicks to play. On the basis of the traditional QoE (Quality of Experience) evaluation, the behavior data of the video user service duration is further introduced. This behavior data directly reflects the preference degree of the video user for the video service, and further reflects the importance degree of the mobile network acceleration requirement.

[0160] The embodiments of the present disclosure determine reasonable threshold values for each KQI indicator through dynamic threshold exploration in combination with the user distribution, so that the system can dynamically adjust the evaluation criteria according to the network environment and user distribution. Compared with the traditional fixed threshold scheme, the dynamic threshold is more flexible and can handle problems such as different wire networks, regions, times, and user requirements, and adapt to the fluctuating changes of the mobile network conditions and user experience.

[0161] In the process of scoring video users, the embodiments of the present disclosure assign weights to the experience indicators and service duration. The weighted scoring mechanism can balance the influence of each indicator, avoid the deviation of a single indicator affecting the overall score, and thus ensure the fairness and accuracy of the recognition result, so that the score can comprehensively and objectively reflect the viewing experience of video users.

[0162] The embodiments of the present disclosure screen out the video users who most need to be preferentially accelerated and improved according to the acceleration guarantee recommendation value, which can ensure that the acceleration resources are allocated to the users with the worst experience, further improve the utilization rate of network resources, and enable the limited network and service resources to more effectively improve the user experience. At the same time, the regular calculation and statistics of the acceleration recommendation value can adapt to the dynamic changes of the network and user behavior, ensuring the timeliness and accuracy of the acceleration strategy.

[0163] Embodiment 3 of the present disclosure also provides a system for processing poor quality of mobile network video users, as Figure 3 shown. The system includes:

[0164] An acquisition module 11 is configured to acquire original user detailed data of cellular mobile network accelerated video users;

[0165] A preprocessing module 12, which is configured to perform data preprocessing on the original user detailed order data;

[0166] An experience index scoring module 13 is configured to obtain experience index data of each video user through the pre-processed original user detail data, so as to perform threshold analysis of the video user experience index and establish a video user experience index scoring table; and

[0167] A service duration scoring module 14 is configured to analyze the video user service duration data through the pre-processed original user detail data and establish a video user service duration scoring table;

[0168] A recommended value determination module 15 is configured to calculate the acceleration guarantee recommended value according to the video user experience index scoring table and the video user service duration scoring table, and obtain the acceleration guarantee recommended value table for the video user;

[0169] The acceleration module 16 is configured to select a video user for network acceleration according to the acceleration guarantee recommendation value table.

[0170] Furthermore, the original user detailed order data includes:

[0171] User-related fields, protocol size class, APPID, effective download rate, XKB startup delay, traffic and time.

[0172] Furthermore, the preprocessing module 12 is specifically configured as follows:

[0173] Extract video user details: match the video user list data in the cellular mobile network according to user-related fields, protocol size class and APPID;

[0174] Data cleaning: Detect video user list data and remove outliers in the data, and use mean filling to supplement the default values ​​in the data;

[0175] Granularity selection and acquisition of video user service duration data: Select a preset time length as the time granularity, count the viewing service duration of each video user in the video user list data after data cleaning, obtain video user service duration data and use it as a behavioral indicator to evaluate video user experience.

[0176] Furthermore,

[0177] The experience indicators include effective download rate and XKB startup delay;

[0178] The experience index scoring module 13 is specifically configured as follows:

[0179] Analyze the effective download rate of video users and the distribution of XKB startup delay through the pre - processed original user detail data, determine the thresholds of the effective download rate and XKB startup delay, and divide the effective download rate and XKB startup delay into multiple scoring intervals respectively;

[0180] Determine the low - rate recommended value interval and high - delay recommended value interval, and determine the scoring values corresponding to each interval of the effective download rate and each interval of XKB startup delay respectively;

[0181] Based on the scoring intervals and scoring value ranges of the effective download rate and XKB startup delay, formulate a scoring table for video user experience indicators.

[0182] Furthermore, the service duration scoring module 14 is specifically set as follows:

[0183] According to actual service experience, divide the service duration of video users into multiple scoring intervals through the pre - processed original user detail data to reflect the activity and continuity of video users watching videos, and set corresponding scoring values for each scoring interval;

[0184] Based on the scoring intervals and scoring values of the service duration, formulate a scoring table for video user service duration.

[0185] Furthermore, the recommended value determination module 15 is specifically set to obtain the video user acceleration guarantee recommended value table through the following steps:

[0186] S1: Preset the weights of the effective download rate and XKB startup delay as a and b respectively, and the weights of video user experience indicators and video user service duration as c and d respectively, and a + b = 1, c + d = 1;

[0187] S2: According to the obtained scoring table of video user experience indicators, calculate the experience value KQI. The specific formula is as follows:

[0188] KQI = LR×a + HL×b;

[0189] Where, LR represents the scoring value corresponding to the interval where the effective download rate of video users is located, and HL represents the scoring value corresponding to the interval where the XKB startup delay of video users is located;

[0190] S3: According to the obtained scoring table of video user service duration, obtain the service duration scoring value, and combine it with the experience value KQI to calculate the acceleration guarantee recommended value RV. The specific formula is as follows:

[0191] RV = KQI×c + T×d;

[0192] Where, T is the scoring value corresponding to the interval where the service duration of video users is located;

[0193] S4: Obtain the video user acceleration guarantee recommendation value table according to the acceleration guarantee recommendation value RV.

[0194] Furthermore, the acceleration module 16 is specifically configured as follows:

[0195] Perform video user slice statistical verification according to the acceleration guarantee recommendation value table, obtain the acceleration guarantee recommendation value threshold, count the acceleration guarantee recommendation times of video users according to the acceleration guarantee recommendation value threshold, and select video users to implement acceleration according to the acceleration guarantee recommendation times of video users.

[0196] Furthermore, the acceleration module 16 is specifically configured as follows:

[0197] Taking the preset time length as the granularity, count the number of times that the acceleration guarantee recommendation value of a video user is greater than or equal to the acceleration guarantee recommendation value threshold within a certain period of time to obtain the acceleration guarantee recommendation times;

[0198] Sort the acceleration guarantee recommendation times of video users within a certain period of time from high to low, and select users whose acceleration guarantee recommendation times are in the top certain proportion of the total number of video users to implement acceleration, so as to ensure that bitrate guarantee is provided preferentially for users with poor experience.

[0199] The mobile network video user quality degradation processing system of the embodiments of the present disclosure is used to implement the mobile network video user quality degradation processing methods in Embodiment 1 and Embodiment 2 of the method, so the description is relatively simple. For specific details, please refer to the relevant descriptions in the previous method embodiments, and will not be elaborated here.

[0200] In addition, as Figure 4 shown, Embodiment 4 of the present disclosure further provides an electronic device, including a memory 100 and a processor 200. A computer program is stored in the memory 100. When the processor 200 runs the computer program stored in the memory 100, the processor 200 executes the above various possible methods.

[0201] Among them, the memory 100 is connected to the processor 200. The memory 100 can adopt flash memory or read-only memory or other memories, and the processor 200 can adopt a central processing unit or a single-chip microcomputer.

[0202] In addition, the embodiments of the present disclosure further provide a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by the processor to perform the above various possible methods.

[0203] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, computer program modules, or other data. The computer-readable storage medium includes, but is not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash memory or other memory technologies, CD-ROM (Compact Disc Read-Only Memory), digital versatile disc (DVD, Digital Video Disc) or other optical disc storage, magnetic cassette, tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0204] It can be understood that the above embodiments are merely exemplary embodiments adopted to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present disclosure, and these modifications and improvements are also considered within the protection scope of the present disclosure.

Claims

1. A method for processing poor quality of mobile network video users, characterized in that: The method comprises: Obtain original user detailed data of cellular mobile network accelerated video users; Performing data preprocessing on the original user detail data; Obtaining experience index data of each video user through the pre-processed original user detail data, so as to perform threshold analysis of the video user experience index and establish a video user experience index scoring table; and, Analyze the service duration data of video users through the pre-processed original user detailed data, and establish a video user service duration score table; Calculate the acceleration guarantee recommendation value based on the video user experience index scoring table and the video user service duration scoring table, and obtain the acceleration guarantee recommendation value table for video users; Select video users for network acceleration based on the acceleration guarantee recommended value table.

2. The method according to claim 1, characterized in that The original user details data includes: User-related fields, protocol size class, application ID APPID, effective download rate, XKB startup delay, traffic and time.

3. The method according to claim 2, characterized in that The data preprocessing of the original user detail list data comprises: Extract video user details: match the video user list data in the cellular mobile network according to user-related fields, protocol size class and APPID; Data cleaning: Detect video user list data and remove outliers in the data, and use mean filling to supplement the default values ​​in the data; Granularity selection and acquisition of video user service duration data: Select a preset time length as the time granularity, count the viewing service duration of each video user in the video user list data after data cleaning, obtain video user service duration data and use it as a behavioral indicator to evaluate video user experience.

4. The method according to claim 1, characterized in that The experience indicators include effective download rate and XKB startup delay; The pre-processed original user detailed data is used to obtain the experience index data of each video user, so as to perform threshold analysis of the video user experience index and establish a video user experience index scoring table, including: The distribution of effective download rate and XKB startup delay of video users is analyzed through the pre-processed original user detailed data, the threshold of effective download rate and XKB startup delay is determined, and the effective download rate and XKB startup delay are divided into multiple scoring intervals respectively; Determine the low rate recommended value interval and the high delay recommended value interval, and respectively determine the score values ​​corresponding to each interval of the effective download rate and each interval of the XKB startup delay; Based on the scoring interval and scoring value range of effective download rate and XKB startup delay, a video user experience index scoring table is developed.

5. The method according to claim 4, characterized in that The pre-processed original user detailed list data is used to analyze the video user service duration data to establish a video user service duration score table, including: According to actual business experience, the service duration of video users is divided into multiple scoring intervals through the pre-processed original user detail data to reflect the activity and continuity of video users in watching videos, and a corresponding scoring value is set for each scoring interval; Based on the scoring range and scoring value of the service time, a video user service time scoring table is formulated.

6. The method according to claim 5, characterized in that The method of calculating the acceleration guarantee recommendation value according to the video user experience index scoring table and the video user service duration scoring table, and obtaining the acceleration guarantee recommendation value table for the video user, includes: S1: The weights of the preset effective download rate and XKB startup delay are a and b respectively, the weights of the video user experience index and the video user service duration are c and d respectively, and a+b=1, c+d=1; S2: Calculate the experience value KQI according to the obtained video user experience index score table. The specific formula is as follows: KQI = LR × a + HL × b; Among them, LR represents the score value corresponding to the interval of the video user's effective download rate, and HL represents the score value corresponding to the interval of the video user's XKB startup delay; S3: According to the obtained video user service duration score table, obtain the service duration score value, and calculate the acceleration guarantee recommended value RV in combination with the experience value KQI. The specific formula is as follows: RV = KQI × c + T × d; Where T is the score corresponding to the interval of the video user's service duration; S4: Obtain a video user acceleration guarantee recommended value table according to the acceleration guarantee recommended value RV.

7. The method according to claim 1, characterized in that The selecting of video users for network acceleration according to the acceleration guarantee recommendation value table includes: Perform statistical verification of video user slices according to the acceleration guarantee recommendation value table, obtain the acceleration guarantee recommendation value threshold, count the number of acceleration guarantee recommendations for video users according to the acceleration guarantee recommendation value threshold, and select video users for acceleration according to the number of acceleration guarantee recommendations for video users.

8. The method according to claim 7, characterized in that The counting of the acceleration guarantee recommendation times of the video user according to the acceleration guarantee recommendation value threshold, and selecting the video user for acceleration according to the acceleration guarantee recommendation times of the video user, comprises: Taking the preset time length as the granularity, count the number of times that the acceleration guarantee recommendation value of the video user is greater than or equal to the acceleration guarantee recommendation value threshold within a certain period of time, and obtain the number of acceleration guarantee recommendations; The number of acceleration guarantee recommendations for video users within a certain period of time is sorted from high to low, and a certain proportion of users whose acceleration guarantee recommendations are before the total number of video users are selected for acceleration to ensure that bitrate guarantee is provided first to users with poor experience.

9. A mobile network video user poor quality processing system, characterized in that: The system comprises: An acquisition module configured to acquire original user detailed list data of cellular mobile network accelerated video users; A preprocessing module, configured to perform data preprocessing on the original user detail order data; An experience index scoring module is configured to obtain experience index data of each video user through pre-processed original user detail data, so as to perform threshold analysis of video user experience index and establish a video user experience index scoring table; and A service duration scoring module is configured to analyze the video user service duration data through the pre-processed original user detail data and establish a video user service duration scoring table; A recommended value determination module, which is configured to calculate the acceleration guarantee recommended value according to the video user experience index scoring table and the video user service duration scoring table, and obtain the acceleration guarantee recommended value table for the video user; The acceleration module is configured to select video users for network acceleration according to the acceleration guarantee recommendation value table.

10. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the method for processing poor user quality of mobile network video according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the method for processing poor user quality of mobile network video according to any one of claims 1 to 8 is implemented.

Images