A user perception evaluation method and device and a computer readable storage medium

Abstract

The application provides a user perception evaluation method, device and computer readable storage medium. The method comprises: obtaining target heavy user data in a target cell, wherein the target heavy user data comprises target network use time data, target network performance data and service domain data; generating a target service behavior slice according to the target network use time data; counting a target number of target service behavior slices that do not meet target services due to the target network performance of the target heavy user within a target service behavior slice duration; and scoring the network in the target cell according to the statistical result and the service domain data to complete user perception evaluation. The method, device and medium can solve the problem that the existing user perception evaluation method only focuses on network side indexes from cells, grids and scenes, resulting in a high rate of user complaints about network quality.

Description

TECHNICAL FIELD

[0001] The present application relates to a user perception evaluation method and device and a computer readable storage medium. BACKGROUND

[0002] At present, the common method for analyzing network quality and user perception in the industry focuses on network side indicators from cells, grids and scenes, and ignores feedback network problems from user perception, which is limited in method. In actual situations, in many cells where network indicators and road tests are not problematic, many users still complain about poor network quality, which leads to the isolation of network indicator optimization and user perception, resulting in a high user complaint rate for network quality. SUMMARY

[0003] The technical problem to be solved by the present application is to solve the above-mentioned deficiencies of the prior art, and to provide a user perception evaluation method, device and computer readable storage medium, which generates a service behavior slice according to the user's use of the network, and scores the network by combining the service domain data and the analysis of the service behavior slice, to solve the problem of high user complaint rate for network quality caused by the existing user perception evaluation method which only focuses on network side indicators from cells, grids and scenes.

[0004] In a first aspect, the present application provides a user perception evaluation method, comprising:

[0005] Obtaining target heavy user data in a target cell, wherein the target heavy user data includes target network usage time data, target network performance data and service domain data;

[0006] Generating a target service behavior slice according to the target network usage time data;

[0007] Counting the target number of unmet targets of the target service behavior slice in which the target network performance does not meet the target service during the duration of the target service behavior slice for the target heavy user;

[0008] Scoring the network in the target cell according to the statistical result and the service domain data to complete the user perception evaluation.

[0009] Preferably, the target heavy user data in the target cell is obtained, comprising:

[0010] Obtaining target user group data in a preset range, including a work parameter table, external data representation XDR user plane data, deep packet inspection DPI data, the target network performance data and the service domain data;

[0011] Determining the target cell location according to the work parameter table;

[0012] acquiring target behavior information of the target user group according to the XDR user plane data and the target cell location, wherein the target behavior information comprises target network usage time data;

[0013] determining the target heavy user of target service according to the target user group behavior information and the deep packet inspection (DPI) data.

[0014] Preferably, the target network usage time data comprises a target application program and x target events corresponding to the target application program, wherein x is an integer greater than 1.

[0015] The target event comprises a target start time and a target end time.

[0016] The target service behavior slice is generated according to the target network usage time data, comprising:

[0017] at least one target application slice is divided according to the target start time and the target end time.

[0018] The slice duration of each target application slice is calculated to generate the target service behavior slice.

[0019] Preferably, the at least one target application slice is divided according to the target start time and the target end time, comprising:

[0020] Step S11, the target events are sorted in chronological order to generate a target event list;

[0021] Step S12, i is initialized to 1;

[0022] Step S13, a target time difference between the target end time of the target event ranked at the i-th position in the target event list and the target start time of the target event ranked at the i+1-th position is calculated.

[0023] Step S14, it is judged whether the target time difference meets a preset threshold, if yes, step S15 is executed, otherwise, the target events ranked at the i-th position and the i+1-th position are merged into a new target event, the target event list is updated and x is reduced by 1, and step S15 is executed.

[0024] Step S15, it is judged whether i is equal to x, if yes, each target event is taken as a target application slice, and the process is ended, otherwise, i is increased by 1, and step S13 is returned.

[0025] Preferably, the network in the target cell is scored according to the statistical result and the service domain data to complete the user perception evaluation, comprising:

[0026] calculating a mean value of the target non-compliance quantity in each target time range within a target period according to the statistical result, the target period comprising a plurality of target time ranges;

[0027] calculating a target non-compliance quantity prediction value according to the mean value of the target non-compliance quantity;

[0028] scoring a network in the target cell according to the target non-compliance quantity prediction value and the service domain data.

[0029] Preferably, the target time range is 1 day, and the target period is 7 days.

[0030] The calculating a mean value of the target non-compliance quantity in each target time range within a target period according to the statistical result, the target period comprising a plurality of target time ranges, satisfies the following formula:

[0031]

[0032] wherein a i is the mean value of the target non-compliance quantity on the i-th day in the target period, a j is the target non-compliance quantity of the j-th hour on the i-th day.

[0033] Preferably, the calculating a target non-compliance quantity prediction value according to the mean value of the target non-compliance quantity satisfies the following formula:

[0034]

[0035] F j = Median(a′ ij )*Base

[0036] wherein F j is the non-compliance quantity prediction value of the j-th hour, Median represents a median, a′ ij is an intermediate coefficient, and Base takes a value satisfying one of the following formulas:

[0037]

[0038]

[0039] In a second aspect, the present application further provides a user perception evaluation device, comprising:

[0040] a data acquisition module configured to acquire target heavy user data in a target cell, wherein the target heavy user data comprises target network usage time data, target network performance data, and service domain data;

[0041] The slice generation module is connected with the data acquisition module and is configured to generate a target service behavior slice according to the target network use time data.

[0042] The quantity statistics module is connected with the slice generation module and is configured to count a target non-meeting quantity of the target service behavior slice in which the target network performance does not meet the target service during the target service behavior slice duration of the target heavy user.

[0043] The perception evaluation module is connected with the quantity statistics module and is configured to score the network in the target cell according to the statistical result and the service domain data to complete the user perception evaluation.

[0044] In a third aspect, the present application further provides a user perception evaluation device, which comprises a memory and a processor, the memory stores a computer program, and the processor is configured to run the computer program to realize the user perception evaluation method in the first aspect.

[0045] In a fourth aspect, the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the user perception evaluation method in the first aspect.

[0046] The user perception evaluation method, device and computer readable storage medium provided by the present application firstly acquire target heavy user data in a target cell, wherein the target heavy user data comprises target network use time data, target network performance data and service domain data, then generate a target service behavior slice according to the target network use time data, count a target non-meeting quantity of the target service behavior slice in which the target network performance does not meet the target service during the target service behavior slice duration of the target heavy user, and finally score the network in the target cell according to the statistical result and the service domain data to complete the user perception evaluation. Since the present application generates a service behavior slice according to the network use condition of the user, scores the network by combining the service domain data and the analysis of the service behavior slice, and solves the problem that the existing user perception evaluation method only focuses on the network side index from the cell, grid and scene, the problem of high network quality complaint rate of the user is solved. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 A flow chart of a user perception evaluation method according to Embodiment 1 of the present application;

[0048] Figure 2 A structure schematic diagram of a user perception evaluation device according to Embodiment 2 of the present application;

[0049] Figure 3 A structure schematic diagram of a user perception evaluation device according to Embodiment 3 of the present application. Detailed Implementation

[0050] To enable those skilled in the art to better understand the technical solution of the present invention, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0051] It is understood that the specific embodiments and accompanying drawings described herein are merely for explaining the invention and are not intended to limit the invention.

[0052] It is understood that, without conflict, the various embodiments and features in the embodiments of the present invention can be combined with each other.

[0053] It is understood that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, while the parts unrelated to the present invention are not shown in the drawings.

[0054] It is understood that each unit or module involved in the embodiments of the present invention may correspond to only one entity structure, or may be composed of multiple entity structures, or multiple units or modules may be integrated into one entity structure.

[0055] It is understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of this invention may occur in a different order than that marked in the accompanying drawings.

[0056] It is understood that the flowcharts and block diagrams of this invention illustrate the possible architecture, functions, and operations of systems, apparatuses, devices, and methods according to various embodiments of this invention. Each block in the flowchart or block diagram may represent a unit, module, program segment, or code, containing executable instructions for implementing the specified function. Furthermore, each block or combination of blocks in the block diagram and flowchart can be implemented using a hardware-based system to achieve the specified function, or using a combination of hardware and computer instructions.

[0057] It is understood that the units and modules involved in the embodiments of the present invention can be implemented by software or by hardware. For example, the units and modules can be located in a processor.

[0058] Example 1:

[0059] This embodiment provides a user perception evaluation method, such as Figure 1 As shown, the method includes:

[0060] Step S101: Obtain target heavy user data within the target cell, wherein the target heavy user data includes target network usage time data, target network performance data, and service domain data.

[0061] In the embodiment, different heavy users can be selected according to different research scenarios, for example, when a campus scenario is selected, the range of the target cell is within the campus, and heavy users of video or game mobile network data services can be selected.

[0062] Optionally, the obtaining of the target heavy user data in the target cell comprises:

[0063] The target user group data in the preset range comprises a work parameter table, XDR (External Data Representation) user plane data, DPI (Deep Packet Inspection) data, the target network performance data, and the service domain data.

[0064] The target cell position is determined according to the work parameter table.

[0065] The target behavior information of the target user group is obtained according to the XDR user plane data and the target cell position, wherein the target behavior information comprises the target network use time data.

[0066] The target heavy user of the target service is determined according to the target user group behavior information and the DPI data.

[0067] In the embodiment, the service identification is a basic function of DPI, which can accurately identify the type of service carried in network traffic, and the service domain data can include user scores, complaints, and application of number portability, and each data of the user corresponds to the number of the user, and the number of the user can be MSISDN (Mobile Station International Subscriber Directory Number).

[0068] Step S102: generating a target service behavior slice according to the target network use time data.

[0069] In the embodiment, the target heavy user is divided into different service types and the time length is calculated, and a target service behavior slice is formed according to mobile network data, start time, and time length.

[0070] Optionally, the target network use time data comprises a target application program and x target events corresponding to the target application program, wherein x is an integer greater than 1.

[0071] The target event comprises a target start time and a target end time.

[0072] The generating of the target service behavior slice according to the target network use time data comprises:

[0073] dividing at least one target application slice according to the target start time and the target end time;

[0074] calculating a slice duration of each target application slice to generate the target service behavior slice.

[0075] Optionally, the dividing at least one target application slice according to the target start time and the target end time comprises:

[0076] Step S11, sorting the target events in time sequence to generate a target event list;

[0077] Step S12, initializing i as 1;

[0078] Step S13, calculating a target time difference between a target end time of the target event ranked at the i-th position in the target event list and a target start time of the target event ranked at the i+1-th position;

[0079] Step S14, judging whether the target time difference satisfies a preset threshold, if yes, executing Step S15, otherwise, merging the target events ranked at the i-th position and the i+1-th position into a new target event, updating the target event list and letting x minus 1, and executing Step S15;

[0080] Step S15, judging whether i is equal to x, if yes, taking each target event as a target application slice, and ending the process, otherwise, letting i plus 1, and returning to Step S13.

[0081] In the embodiment, the grouping can be performed according to user numbers and application programs, the grouping can be performed according to application program start times, a time difference of adjacent records can be obtained by subtracting a start time of a next record from an end time of a previous record, if the time difference is greater than a preset threshold, it is considered that a user does not continuously use a same application program, at this time, a mark field can be added between two groups of records, and the mark field can be assigned a value of 1, the preset threshold can be set as 300 seconds, the mark field represents a frequency of use of a user in a same application program group, and multiple records in a same group are divided into multiple slices.

[0082] After the slices are divided, an encoding of a slice can be generated according to an accumulated value of a mark field, and a duration of the slice can be obtained by subtracting a minimum start time from a maximum end time in a same slice.

[0083] Step S103: counting a target non-meeting quantity of the target service behavior slice in which the target network performance does not meet the target service during a target service behavior slice duration of the target heavy user.

[0084] In the embodiment, different network performance thresholds can be set for different service types, and whether each slice meets the threshold is determined in sequence, for example, the service interaction delay of the target service behavior slice corresponding to the mobile game and navigation out-of-home service is less than 500 ms, which can be considered as meeting the threshold.

[0085] Step S104: scoring the network in the target cell according to the statistical result and the service domain data to complete the user perception evaluation.

[0086] In the embodiment, different weights can be set for the service domain data and the statistical result for scoring, for example, the weight of the target unmet quantity is 70%, and the user score, complaint and application for number portability data in the service domain data accounts for 10% of the weight respectively, and finally the score can be normalized to the interval of 0 to 100.

[0087] Optionally, the scoring the network in the target cell according to the statistical result and the service domain data to complete the user perception evaluation comprises:

[0088] According to the statistical result, the average value of the target unmet quantity in each target time range in a target period containing multiple target time ranges is calculated;

[0089] According to the average value of the target unmet quantity, a target unmet quantity prediction value is calculated;

[0090] According to the target unmet quantity prediction value and the service data, the network in the target cell is scored.

[0091] Optionally, the target time range is 1 day, and the target period is 7 days;

[0092] According to the statistical result, the average value of the target unmet quantity in each target time range in a target period containing multiple target time ranges is calculated, which satisfies the following formula:

[0093]

[0094] Wherein, a i is the average value of the target unmet quantity on the i-th day in the target period, and a j is the target unmet quantity of the j-th hour on the i-th day.

[0095] Optionally, the target unmet quantity prediction value calculated according to the average value of the target unmet quantity satisfies the following formula:

[0096]

[0097] F j = Median(a′ ijBase

[0098] wherein F j is the prediction of the number of out-of-compliance for the jth hour, Median represents the median, a′ ij is the intermediate coefficient, and Base takes a value satisfying one of the following formulas:

[0099]

[0100]

[0101] In the present embodiment, the median of F j may be selected to represent the out-of-compliance slice condition of the target cell, because the median has better robustness.

[0102] In one specific embodiment, the user perception evaluation method can include the following steps:

[0103] S1. Obtain heavy user data:

[0104] Obtain an ECI (E-UTRAN Cell Identifier) list according to the filtering condition. For example, the present embodiment selects a campus scenario as the research object, and obtains the ECIs involved in the latitude and longitude range of multiple schools from the engineering parameter table.

[0105] Associate XDR user plane HTTP (Hyper text transfer protocol) data according to the ECI list, and form a user behavior information detail table, including MSISDN, APP (application) name, date, start time, end time, time difference, and total traffic. The repeated data in the user behavior information detail table can be deleted by taking the MSISDN field as the primary key.

[0106] The research example is a heavy user of mobile network data services of video or game. Because video in streaming media and game in time-sensitive services are typical, the requirements of users on network continuity, real-time performance, and timing can be better reflected. The number of users, traffic, and duration are calculated according to the records of video and game in the user behavior detail table. The users with the top 20% of video and game usage traffic or duration are taken as heavy users to form heavy user lists of the two types of services. Because the top 20% of users' traffic or duration can account for more than 90% of the total traffic or duration, the reduction of the number of users will also improve the efficiency of association of different data tables.

[0107] The heavy user data can be shown in the following table:

[0108]

[0109] Heavy user data also includes B (business) domain data

[0110] Obtain the information of the user involved in the cell in the B domain, including the user's score for the network, whether to complain, whether to send a number transfer network application, etc.

[0111] The B domain data is shown in the following table:

[0112]

[0113] S2. Form time slices based on user dynamic behavior data:

[0114] S21. Calculate the time difference Δt:

[0115] According to the user number, APP, group the records in the group according to Start time, and calculate the time difference Δt between the Start time of the next record and the End time of the previous record.

[0116] S22. Calculate the slice label Session Tag:

[0117] If the time interval Δt is greater than the threshold value 300s, it is considered that the user does not continuously use the same APP, at this time the Session Tag field is added and assigned a value of 1, the Session Tag field represents the frequency of the user's use in the same APP group, and the records in the same group are divided into multiple slices.

[0118] S23. Calculate the slice number Session Id:

[0119] The Session Id field is formed by accumulating the Session Tag field, after grouping according to date, user number and APP, the records in the group are sorted according to Start time, and the Session Id is obtained by accumulating the Session Tag to represent the coding of the slice.

[0120] S24. Calculate the slice duration Session T:

[0121] In the same Session Id number, the maximum End time minus the minimum Start time, the duration of the slice Session T is obtained.

[0122] After steps S21 to S24, the user mobile network service behavior slice is formed.

[0123] S3. Label the network perception of the slice combined with user network performance data

[0124] S31. Calculate the slice downlink rate, interaction delay and waiting delay. According to different service types, set the indicators and thresholds of network performance, and label each slice as "up to standard" or "not up to standard" in turn.

[0125] The performance index calculation method and threshold corresponding to part of the service type are shown in the following table:

[0126]

[0127] S4. Cell out-of-standard prediction: use periodic factor method to predict the number of out-of-standard slices in the cell range

[0128] S41. Count the number of out-of-standard slices per day for each user.

[0129] S42. Calculate the number of out-of-standard slices per hour for n users in the cell range, which satisfies the following formula:

[0130]

[0131] S43. Predict the number of out-of-standard slices in the cell range:

[0132] Accumulate the out-of-standard slice data of the cell for 7 days in units of hours. For a certain cell, a ij i is a positive integer from 1 to 7, representing the day; j is a positive integer from 0 to 23, representing the hour. The average value of the out-of-standard slices of the cell per day a i satisfies the following formula:

[0133]

[0134] Select the median M ij (j=0, …, 23) from a′ j , and obtain a set of robust median factors not affected by extreme values:

[0135]

[0136] M=Median(a′ ij )

[0137] F j =M*Base

[0138] F j as the prediction of the number of out-of-standard slices per hour on the eighth day. The calculation method of Base satisfies the following formula:

[0139]

[0140]

[0141] Specifically, if the target date to be predicted is a day of the week, the average of the 7 weekdays (Base1) is selected. If the target date to be predicted is a weekend, the average of the 7 weekend days (Base2) is selected.

[0142] Take F j The median F represents the substandard slices in the cell, because the median is robust.

[0143] S5. Comprehensive Evaluation: The predicted number of substandard slices and B-domain data are comprehensively evaluated to form a cell network perception evaluation algorithm, which is used to comprehensively evaluate the cells involved.

[0144] The weights and scoring rules for each assessment item are shown in the table below:

[0145]

[0146]

[0147] A perception score between [0, 100] can be obtained for each cell. This provides a user perception ranking of the cell network, helping professionals to focus on cells with poor perception and low rankings.

[0148] In this embodiment, all containers are stored in the cloud server as image files. Each container is independent of the others. When a user needs to use a specific operating system to run a specific application, they can start the container corresponding to the application.

[0149] The user perception assessment method provided in this invention first acquires target heavy user data within a target cell, including target network usage time data, target network performance data, and service domain data. Then, it generates target service behavior slices based on the target network usage time data. It then counts the number of target service behavior slices where the target network performance does not meet the target service requirements during the duration of the target service behavior slice for each target heavy user. Finally, it scores the network within the target cell based on the statistical results and the service domain data to complete the user perception assessment. Because this invention generates service behavior slices based on user network usage and scores the network by combining service domain data and analysis of the service behavior slices, it solves the problem of high user complaint rates regarding network quality caused by existing user perception assessment methods that only focus on network-side indicators from the cell, grid, and scenario levels.

[0150] Example 2:

[0151] like Figure 2 As shown, this embodiment provides a user perception assessment device for performing the above-described user perception assessment method, including:

[0152] The data acquisition module 201 is configured to acquire target heavy user data in a target cell, wherein the target heavy user data comprises target network usage time data, target network performance data, and service domain data.

[0153] The slice generation module 202 is connected with the data acquisition module 201 and configured to generate a target service behavior slice according to the target network usage time data.

[0154] The quantity statistics module 203 is connected with the slice generation module 202 and configured to count a target non-compliance quantity of the target heavy user in the target service behavior slice, which is not satisfied with the target service performance of the target service behavior slice in a target service behavior slice duration.

[0155] The perception evaluation module 204 is connected with the quantity statistics module 203 and configured to score the network in the target cell according to the counting result and the service domain data to complete user perception evaluation.

[0156] Preferably, the data acquisition module 201 specifically comprises:

[0157] The acquisition unit is configured to acquire target user group data in a preset range, including a work parameter table, XDR user plane data, DPI data, the target network performance data, and the service domain data.

[0158] The work parameter table unit is configured to determine the target cell position according to the work parameter table.

[0159] The behavior unit is configured to acquire target behavior information of the target user group according to the XDR user plane data and the target cell position, wherein the target behavior information comprises the target network usage time data.

[0160] The DPI unit is configured to determine the target heavy user of the target service according to the target user group behavior information and the DPI data.

[0161] Preferably, the target network usage time data comprises a target application program and x target events corresponding to the target application program, wherein x is an integer greater than 1.

[0162] The target event comprises a target start time and a target end time.

[0163] The slice generation module 202 specifically comprises:

[0164] The slice unit is configured to divide at least one target application slice according to the target start time and the target end time.

[0165] a computing unit configured to calculate a slice duration of each target application slice to generate the target service behavior slice.

[0166] Preferably, the slicing unit specifically comprises:

[0167] a sorting unit configured to sort the target events in chronological order to generate a target event list;

[0168] an initializing unit configured to initialize i as 1;

[0169] a time difference unit configured to calculate a target time difference between a target end time of the target event ranked at the i-th position in the target event list and a target start time of the target event ranked at the i+1-th position;

[0170] a first judging unit configured to judge whether the target time difference meets a preset threshold, if yes, execute a second judging unit, otherwise, merge the target events ranked at the i-th position and the i+1-th position into a new target event, update the target event list and let x minus 1, and execute the second judging unit;

[0171] the second judging unit configured to judge whether i is equal to x, if yes, take each target event as a target application slice, and end the process, otherwise, let i plus 1, and return to the time difference unit.

[0172] Preferably, the perception evaluation module 204 specifically comprises:

[0173] a mean unit configured to calculate a mean value of the target substandard number in each target time range within a target period containing multiple target time ranges according to the statistical result;

[0174] a prediction unit configured to calculate a target substandard number prediction value according to the mean value of the target substandard number;

[0175] a scoring unit configured to score the network in the target cell according to the target substandard number prediction value and the service data.

[0176] Preferably, the target time range is 1 day, and the target period is 7 days.

[0177] The mean unit satisfies the following formula:

[0178]

[0179] wherein, a i is the mean value of the target substandard number on the i-th day in the target period, and a j is the target substandard number of the j-th hour on the i-th day.

[0180] Preferably, the prediction unit satisfies the following formula:

[0181]

[0182] F j = Median(a' ij ) * Base

[0183] Wherein, F j is the jth hour of the number of substandard prediction value, Median represents the median, a' ij is the intermediate coefficient, Base value satisfies one of the following formulas:

[0184]

[0185]

[0186] Embodiment 3:

[0187] As Figure 3 shown, the embodiment provides a user perception evaluation device for executing the above-mentioned user perception evaluation method, comprising a memory 301 and a processor 302, the memory 301 stores a computer program, and the processor 301 is configured to run the computer program to execute the user perception evaluation method in embodiment 1.

[0188] Wherein, the memory 301 is connected with the processor 301, the memory 301 can adopt flash memory or read-only memory or other storage, and the processor 302 can adopt central processing unit or single-chip microcomputer.

[0189] Embodiment 4:

[0190] The embodiment provides a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize the user perception evaluation method in the above-mentioned embodiment 1.

[0191] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storage of 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 technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disc (DVD) or other optical disc storage, magnetic cassettes, magnetic tapes, 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.

[0192] The user perception evaluation device and the computer readable storage medium provided by embodiments 2 to 4 first acquire target heavy user data in a target cell, wherein the target heavy user data includes target network usage time data, target network performance data and service domain data, then generate a target service behavior slice according to the target network usage time data, count the target non-compliance number of the target service behavior slice whose target network performance does not meet the target service in the target service behavior slice duration, and finally score the network in the target cell according to the statistical result and the service domain data to complete the user perception evaluation. Since the application generates a service behavior slice according to the user's network usage, scores the network in combination with the service domain data and the analysis of the service behavior slice, the problem of high user network quality complaint rate caused by the existing user perception evaluation method focusing on network side indicators from cells, grids and scenes is solved.

[0193] It can be understood that the above embodiments are only exemplary embodiments adopted for illustrating the principles of the present application, and the present application is not limited thereto. Various modifications and improvements can be made by those skilled in the art without departing from the spirit and essence of the present application, and these modifications and improvements are also considered to be within the protection scope of the present application.

Claims

1. A user perception assessment method, characterized by, The method comprises: acquiring target heavy user data in a target cell, wherein the target heavy user data comprises target network usage time data, target network performance data, and service domain data; generating a target service behavior slice according to the target network usage time data; counting a target non-meeting quantity of the target heavy user in the target service behavior slice duration, which does not meet the target service behavior slice of the target network performance; scoring the network in the target cell according to the counting result and the service domain data to complete user perception evaluation; the target network usage time data comprises a target application program and x target events corresponding to the target application program, wherein x is an integer greater than 1; the target event comprises a target start time and a target end time; the generating of the target service behavior slice according to the target network usage time data comprises: dividing at least one target application slice according to the target start time and the target end time; calculating the slice duration of each target application slice to generate the target service behavior slice; the dividing of at least one target application slice according to the target start time and the target end time comprises: step S11, sorting the target events in time sequence to generate a target event list; step S12, initializing i as 1; step S13, calculating a target time difference between the target end time of the target event ranked in the i th position in the target event list and the target start time of the target event ranked in the i+1 th position; step S14, judging whether the target time difference meets a preset threshold, if yes, executing step S15, otherwise, merging the target events ranked in the i th position and the i+1 th position into a new target event, updating the target event list and letting x decrease by 1, and executing step S15; step S15, judging whether i is equal to x, if yes, taking each target event as a target application slice, and ending the process, otherwise, letting i increase by 1 and returning to step S13.

2. The user perception assessment method of claim 1, wherein the acquiring of the target heavy user data in the target cell comprises: acquiring target user group data in a preset range, including a work parameter table, XDR user plane data, DPI data, the target network performance data, and the service domain data; determining the target cell location according to the work parameter table; acquiring target behavior information of the target user group according to the XDR user plane data and the target cell location, wherein the target behavior information comprises the target network usage time data; determining the target heavy user of the target service according to the target user group behavior information and the DPI data.

3. The user perception assessment method of claim 1, wherein, the scoring of the network in the target cell according to the counting result and the service domain data to complete user perception evaluation comprises: calculating the mean value of the target non-meeting quantity in each target time range in a target period containing multiple target time ranges according to the counting result; calculating a target non-meeting quantity prediction value according to the mean value of the target non-meeting quantity. Score the network in the target cell according to the target non-compliance quantity prediction value and the service domain data.

4. The user perception assessment method of claim 3, wherein, The target time range is 1 day, and the target period is 7 days. The target non-compliance quantity prediction value is calculated according to the target non-compliance quantity mean value, and the following formula is met: wherein a i is the mean of the target non-compliance quantity for day i in the target period, a ij is the target non-compliance quantity for hour j on day i.

5. The user perception assessment method of claim 4, wherein, The target non-compliance quantity prediction value is calculated according to the target non-compliance quantity mean value, and the following formula is met: F j = Median(a′ ij )*Base Wherein, F j is the prediction value of the non-compliance quantity of the jth hour, Median represents the median, a′ ij is the intermediate coefficient, i is a positive integer from 1 to 7, representing days, j is a positive integer from 0 to 23, representing hours, and Base takes a value satisfying one of the following formulas:

6. A user perception assessment apparatus, characterized by, Comprise: A data acquisition module is configured to acquire target heavy user data in a target cell, wherein the target heavy user data comprises target network usage time data, target network performance data, and service domain data; A slice generation module is connected with the data acquisition module and configured to generate target service behavior slices according to the target network usage time data; A quantity statistics module is connected with the slice generation module and configured to count a target non-compliance quantity of the target network performance not meeting the target service in the target service behavior slices during the target service behavior slice duration of the target heavy user; A perception evaluation module is connected with the quantity statistics module and configured to score the network in the target cell according to the statistical result and the service domain data to complete user perception evaluation; The target network usage time data comprises a target application program and x target events corresponding to the target application program, wherein x is an integer greater than 1; The target event comprises a target start time and a target end time; The slice generation module specifically comprises: A slice unit is configured to divide at least one target application slice according to the target start time and the target end time; A calculation unit is configured to calculate the slice duration of each target application slice to generate the target service behavior slice; The slice unit specifically comprises: An ordering unit is configured to sort the target events in chronological order to generate a target event list; An initialization unit is configured to initialize i as 1; A time difference unit is configured to calculate a target time difference between the target end time of the target event ranked in the i-th position and the target start time of the target event ranked in the i+1-th position in the target event list; A first judgment unit is configured to judge whether the target time difference meets a preset threshold value, if yes, execute a second judgment unit, otherwise, merge the target events ranked in the i-th position and the i+1-th position into a new target event, update the target event list and let x minus 1, and execute the second judgment unit; The second judgment unit is configured to judge whether i is equal to x, if yes, take each target event as a target application slice, and end the process, otherwise, let i plus 1, and return to the time difference unit.

7. A user perception assessment apparatus, characterized by, The computer readable storage medium stores a computer program, and the processor is configured to run the computer program to realize the user perception evaluation method of any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the processor is configured to run the computer program to realize the user perception evaluation method of any one of claims 1-5.

Images