Network quality perception prediction method and apparatus, and storage medium

Abstract

The embodiments of the present application belong to the field of communication networks. Provided are a network quality perception prediction method and apparatus, and a storage medium. The method comprises: acquiring target log data of a user within a preset time period; classifying the target log data on the basis of a service type to which the target log data belongs, so as to obtain target service data corresponding to at least one preset service label; respectively inputting, into a first target classifier corresponding to each preset service labels, target service data corresponding to each preset service label, so as to obtain a first classification result; and inputting the first classification result into a second target classifier, so as to obtain a network quality perception prediction result which is determined by the second target classifier on the basis of the first classification result.

Description

Network quality perception prediction method, device and storage medium

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to the Chinese patent application filed with the China Patent Office on September 26, 2023, with application number 202311253978.4 and invention name “Network Quality Perception Prediction Method, Device and Storage Medium”. The entire contents of the Chinese patent application are incorporated herein by reference. Technical Field

[0003] The present application relates to the field of communication network technology, and in particular to a network quality perception prediction method, device and storage medium. Background Art

[0004] By predicting users' network quality perception, identifying users who perceive poor network quality, and proactively addressing potential complaints, network service quality can be effectively improved. However, different users perceive network quality differently depending on the services they use. For example, on the same network, users may perceive network quality differently when watching live broadcasts or reading novels. Existing methods for predicting network quality perception are generally limited and struggle to accurately predict network quality perception in diverse scenarios.

[0005] Summary of the Invention

[0006] In a first aspect, an embodiment of the present application provides a network quality perception prediction method, including: obtaining target log data of a user within a preset time period; classifying the target log data according to the business type to which the target log data belongs, and obtaining target business data corresponding to at least one preset business tag; inputting the target business data corresponding to each of the preset business tags into a first target classifier corresponding to each preset business tag, and obtaining a first classification result; inputting each of the first classification results into a second target classifier, and obtaining a network quality perception prediction result determined by the second target classifier based on the first classification result.

[0007] In the second aspect, an embodiment of the present application also provides a network quality perception prediction device, which includes a processor, a memory, a computer program stored on the memory and executable by the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein when the computer program is executed by the processor, the steps of any network quality perception prediction method provided in the embodiment of the present application are implemented.

[0008] In a third aspect, an embodiment of the present application further provides a storage medium for computer-readable storage, characterized in that the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any network quality perception prediction method provided in an embodiment of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] FIG1 is a schematic diagram of a flow chart of the steps of a network quality perception prediction method provided by an embodiment of the present application;

[0010] FIG2 is a schematic flow chart of the sub-steps of the method for obtaining target log data in FIG1 ;

[0011] FIG3 is a schematic flow chart of the sub-steps of the method for obtaining the network quality perception prediction result in FIG1 ;

[0012] FIG4 is a schematic diagram of a sub-step flow chart of a model training method provided in another embodiment of the present application;

[0013] FIG5 is a schematic diagram of a scenario for implementing the network quality perception prediction method provided in an embodiment of the present application;

[0014] FIG6 is a schematic block diagram of the structure of a network quality perception prediction device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0016] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0017] It should be understood that the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0018] The present invention provides a method, apparatus, and storage medium for predicting network quality perception. The method can be applied to a mobile terminal, such as a mobile phone, tablet computer, laptop computer, desktop computer, personal digital assistant, or wearable device.

[0019] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0020] In related technologies, to improve the quality of communication network services, it is necessary to predict users' network quality perceptions and determine whether they are satisfied with the network quality under the current communication conditions, that is, whether they perceive poor quality. However, related network quality perception prediction methods typically directly train machine learning models, deep learning models, or other pre-set models using network logs. This ignores the differences in perception caused by different users' varying demands for network quality. For example, online game players are generally more sensitive to network quality. Online game players have a strong sense of competition and teamwork. Lag or disconnections caused by poor network quality can significantly impact their perception of network quality, making them more likely to perceive poor quality. Conversely, text readers are less sensitive to network quality. The content they browse places less demanding demands on the network, and because text readers are relatively well-educated, they can understand fluctuations in network quality within a reasonable range.

[0021] Please refer to Figure 1, which is a schematic flow chart of the steps of a network quality perception prediction method provided by an embodiment of the present application.

[0022] As shown in FIG1 , the network quality perception prediction method includes steps S101 to S104 .

[0023] Step S101: Obtain target log data of a user within a preset time period.

[0024] For example, a Deep Packet Inspection (DPI) probe is used to obtain target log data within a preset time period. Specifically, the DPI probe can be deployed on an Optical Line Terminal (OLT) device, such as an uplink port on the OLT device, i.e., the port that connects the OLT device to the upper-layer switch. Of course, this is not limited to this; the DPI probe can also be deployed on other carrier-based bearer equipment, which is not a limitation here.

[0025] Exemplarily, the target log data may be XDR (X Detailed Record) data. XDR data refers to a detailed record of signaling and services generated after processing based on the full data for use by network fault analysis systems, signaling monitoring / analysis platforms, and signaling applications.

[0026] Please refer to FIG. 2 , which is a schematic flow chart of sub-steps of the method for obtaining target log data in FIG. 1 .

[0027] As shown in Figure 2, in some embodiments, step S101 obtains the target log data of the user within a preset time period, including: step S1011, obtaining the original traffic data in the carrier device; step S1012, parsing the original traffic data to obtain Internet access log data; step S1013, filtering the Internet access log data according to preset filtering rules to obtain the target log data.

[0028] For example, the original traffic data in the bearer device of the home broadband user is mirrored and the original traffic data is parsed to obtain the Internet log data. The Internet log data may include the user's Internet access time information field, service type field, traffic size field, delay information field, packet loss rate, retransmission rate and other information.

[0029] In some embodiments, filtering the Internet log data according to preset filtering rules to obtain the target log data includes: filtering preset log data in the Internet log data, and the preset log data includes: heartbeat signal data, small traffic data with a traffic value less than a preset value, and overseas traffic data.

[0030] For example, the Internet access log data includes all log data during the user's Internet access process, among which there may be some log data that is not of reference value, such as heartbeat signal data generated by the user device sending and receiving heartbeat signals at regular time intervals, small traffic data generated due to accidental factors, and overseas traffic data that is not controlled by the operator, so as to obtain target log data that is more representative of the user's perception.

[0031] Among them, the heartbeat signal data and overseas traffic data can be identified based on the information recorded in the Internet log data. The preset value for judging the small traffic data can be set according to actual needs and is not limited here.

[0032] Step S102: Classify the target log data according to the business type to which the target log data belongs, and obtain target business data corresponding to at least one preset business tag.

[0033] In order to make the network quality perception prediction method provided in the embodiment of the present application applicable to various network scenarios, the target log data is classified to obtain the target business data corresponding to each preset business tag.

[0034] For example, the number and type of service tags can be set according to actual needs. For example, service tags may include: live broadcast users, online course users, online game users, short video users, text reading users, programmer users, etc., without limitation.

[0035] For example, by merging target log data with the same business label, the computational complexity in the model training and network quality perception prediction process is reduced, and business characteristics are given to the target log data, thereby improving the interpretability of the prediction process.

[0036] In some embodiments, the target log data is classified according to the business type to which the target log data belongs to obtain target business data corresponding to at least one preset business tag, including: clustering the target log data according to the business type field in the target log data to obtain target business data corresponding to each preset business tag.

[0037] Exemplarily, the target log data records a service type field for indicating the service to which the target log data corresponds, for example, indicating that the service being used by the user when the target log data is recorded is WeChat, Taobao, Weibo, etc.

[0038] For example, the target log data containing each business type field can be clustered according to the correspondence between the pre-set business type field and the pre-set business tag. Of course, this is not limited to this. A clustering algorithm model can also be pre-trained to cluster the target log data, which is not limited here.

[0039] In some embodiments, the target log data is classified according to the business type to which the target log data belongs, and target business data corresponding to at least one preset business tag is obtained, including: according to the preset business tag corresponding to the target log data, the target business data of each preset business tag is counted, and the target business data includes: at least one of uplink traffic data, downlink traffic data, duration data, log quantity data, delay data, and retransmission data.

[0040] Exemplarily, statistics are collected on the target log data belonging to the same preset business tag to determine the various indicators of the target business data under the preset business tag. Specifically, the uplink traffic data in the target log data belonging to the same preset business tag are added together, and so on, to obtain the uplink traffic data, downlink traffic data, duration data, and log quantity data corresponding to the target business data under the preset business tag, and at the same time calculate the mean, variance, and quantile of the delay data and retransmission data; thereby determining the statistical indicators of one or more of the downlink traffic data, duration data, log quantity data, delay data, and retransmission data corresponding to the target business data under the preset business tag.

[0041] Step S103: input the target service data corresponding to each preset service tag into the first target classifier corresponding to each preset service tag to obtain a first classification result.

[0042] Exemplarily, multiple first target classifiers corresponding to the number of preset business tags are set, which are used to process the target business data of each preset business tag respectively, so that each first target classifier can process the target business data corresponding to each preset business tag in a targeted manner, thereby improving the accuracy of network quality perception prediction.

[0043] In some embodiments, the target business data corresponding to each preset business tag is respectively input into the first target classifier corresponding to each preset business tag to obtain a first classification result, including: when there is no corresponding target business data for the preset business tag, the preset value is input into the first target classifier as the target business data corresponding to the preset business tag.

[0044] Exemplarily, the number of preset business tags corresponds to the number of first target classifiers, but the target log data collected for the same home broadband user may not include log data of all business types. Therefore, there may be a situation where there is no corresponding target business data for a preset business tag.

[0045] Exemplarily, for a preset service tag for which target service data does not exist, the target service data is supplemented with a preset value and input into the corresponding first target classifier.

[0046] Step S104: input each of the first classification results into a second target classifier to obtain a network quality perception prediction result determined by the second target classifier according to the first classification results.

[0047] Exemplarily, a preliminary classification result is obtained by a first target classifier, and then a final network quality perception prediction result is determined by a second target classifier based on the first classification result, thereby improving the accuracy of the network quality perception prediction method.

[0048] Please refer to FIG3 , which is a flowchart of the sub-steps of the method for obtaining the network quality perception prediction result in FIG1 .

[0049] As shown in Figure 3, in some embodiments, step S104 inputs each of the first classification results into the second target classifier to obtain a network quality perception prediction result determined by the second target classifier based on the first classification result, including: step S1041, splicing the first classification results to obtain a target input result; step S1042, inputting the target input result into the second target classifier to obtain the network quality perception prediction result, wherein the network quality perception prediction result includes: positive perception result and negative perception result.

[0050] Exemplarily, step S103 obtains the first classification results corresponding to the number of first target classifiers. The first classification results can be in the form of an array, for example. In order to input multiple first classification results into a preset single second target classifier, the multiple first classification results can be spliced ​​to obtain the target input result, and the target input result is input into the second target classifier.

[0051] Exemplarily, the network quality perception prediction results include at least a positive perception result and a negative perception result. It is understood that a positive perception result is a non-poor perception result, and a negative perception result is a poor perception result. Of course, the network quality perception prediction results are not limited to this. For example, a negative perception result can include multiple levels of negative perception results.

[0052] Please refer to Figure 4, which is a schematic diagram of the sub-step flow of the model training method provided in another embodiment of the present application.

[0053] As shown in Figure 4, in some embodiments, the method also includes: inputting sample log data into a first preset classifier, and obtaining a sample perception prediction result of a second preset classifier based on the sample log data; adjusting the parameters of the first preset classifier and the second preset classifier according to the preset network quality perception label and the sample perception prediction result until the first target classifier and the second target classifier are obtained; wherein, the first preset classifier includes: at least one of: a random forest classifier, an adaptive boosting classifier, and a support vector machine classifier; the second preset classifier includes: a logistic regression classifier.

[0054] Exemplarily, the network quality perception prediction method provided in an embodiment of the present application further includes training a first preset classifier and a second preset classifier, adjusting parameters of the first preset classifier and the second preset classifier, and obtaining a first target classifier and a second target classifier. The multiple first preset classifiers are classifiers of the same type, for example, all of the multiple first preset classifiers are random forest classifiers.

[0055] Exemplarily, by comparing the sample perception prediction results obtained by the sample log data through the first preset classifier and the second preset classifier, as well as the preset network quality perception label corresponding to the sample log data, the parameters of the first preset classifier and the second preset classifier are adjusted until the loss value of the first preset classifier and the second preset classifier is less than the preset loss value, thereby obtaining the first target classifier and the second target classifier.

[0056] Among them, the sub-steps of the model training method of step S201-step S202 can be executed before step S101 or after step S104. Of course, it is not limited to this. Steps S201-step S202 can also be executed while executing steps S101-step S104. There is no limitation here.

[0057] In some embodiments, before adjusting the parameters of the first preset classifier and the second preset classifier according to the preset network quality perception label and the sample perception prediction result, the method further includes: obtaining the network quality perception label fed back by the user.

[0058] For example, the network quality perception tag corresponding to the sample log data is determined based on actual user feedback corresponding to the sample log data. Specifically, user feedback can be obtained by calling customer service or through questionnaires, etc., without limitation.

[0059] Please refer to FIG. 5 , which is a schematic diagram of a scenario for implementing the network quality perception prediction method provided in an embodiment of the present application.

[0060] As shown in Figure 5, the target log data is classified according to the business type field to obtain the target business data corresponding to each preset business label, and each target business data is input into the corresponding first target classifier to obtain the first classification result, and then multiple first classification results are input into the second target classifier to obtain the network quality perception prediction result.

[0061] The network quality perception prediction method provided in the above embodiment obtains the user's target log data within a preset time period; classifies the target log data according to the business type to which the target log data belongs, and obtains target business data corresponding to at least one preset business tag; inputs the target business data corresponding to each preset business tag into the first target classifier corresponding to each preset business tag to obtain a first classification result; and inputs each first classification result into a second target classifier to obtain a network quality perception prediction result determined by the second target classifier based on the first classification result. Since the target log data is classified according to business type and first target classifiers are set separately for target business data of different business types, the target log data can be processed for different scenarios, thereby improving the accuracy of network quality perception prediction.

[0062] Please refer to FIG. 6 , which is a schematic block diagram of the structure of a network quality perception prediction device provided in an embodiment of the present application.

[0063] As shown in FIG6 , the network quality perception prediction device 300 includes a processor 301 and a memory 302 . The processor 301 and the memory 302 are connected via a bus 303 , which is, for example, an I2C (Inter-integrated Circuit) bus.

[0064] Specifically, the processor 301 is used to provide computing and control capabilities to support the operation of the entire network quality perception and prediction device. The processor 301 can be a central processing unit (CPU). The processor 301 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0065] Specifically, the memory 302 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0066] Those skilled in the art will understand that the structure shown in Figure 6 is merely a block diagram of a partial structure related to the embodiment of the present application, and does not constitute a limitation on the network quality perception prediction device to which the embodiment of the present application is applied. The specific network quality perception prediction device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0067] The processor is configured to run a computer program stored in a memory, and implement any one of the network quality perception prediction methods provided in the embodiments of the present application when executing the computer program.

[0068] In one embodiment, the processor is configured to run a computer program stored in the memory, and implement the following steps when executing the computer program:

[0069] Obtain the user's target log data within a preset time period;

[0070] Classifying the target log data according to the business type to which the target log data belongs, and obtaining target business data corresponding to at least one preset business tag;

[0071] Inputting the target service data corresponding to each of the preset service tags into the first target classifier corresponding to each of the preset service tags to obtain a first classification result;

[0072] Each of the first classification results is input into a second target classifier to obtain a network quality perception prediction result determined by the second target classifier according to the first classification results.

[0073] In one embodiment, when obtaining target log data of a user within a preset time period, the processor is configured to:

[0074] Obtain raw traffic data from the bearer device;

[0075] Parsing the original traffic data to obtain Internet access log data;

[0076] The Internet access log data is filtered according to a preset filtering rule to obtain the target log data.

[0077] In one embodiment, when filtering the Internet access log data according to a preset filtering rule to obtain the target log data, the processor is configured to implement:

[0078] Preset log data in the Internet access log data is filtered, and the preset log data includes: heartbeat signal data, small flow data with a flow value less than a preset value, and overseas flow data.

[0079] In one embodiment, when the processor classifies the target log data according to the business type to which the target log data belongs and obtains the target business data corresponding to at least one preset business tag, it is configured to implement:

[0080] The target log data is clustered according to the service type field in the target log data to obtain target service data corresponding to each preset service tag.

[0081] In one embodiment, when the processor classifies the target log data according to the business type to which the target log data belongs and obtains the target business data corresponding to at least one preset business tag, it is configured to implement:

[0082] According to the preset service tags corresponding to the target log data, the target service data of each preset service tag is counted, and the target service data includes: at least one of uplink traffic data, downlink traffic data, duration data, log quantity data, delay data, and retransmission data.

[0083] In one embodiment, when the processor implements inputting the target service data corresponding to each preset service tag into the first target classifier corresponding to each preset service tag to obtain the first classification result, it is configured to implement:

[0084] In the case that the preset service tag does not have corresponding target service data, a preset value is input into the first target classifier as the target service data corresponding to the preset service tag.

[0085] In one embodiment, when inputting each first classification result into a second target classifier to obtain a network quality perception prediction result determined by the second target classifier based on the first classification result, the processor is configured to implement:

[0086] splicing the first classification results to obtain a target input result;

[0087] The target input result is input into the second target classifier to obtain the network quality perception prediction result, wherein the network quality perception prediction result includes: a positive perception result and a negative perception result.

[0088] In one embodiment, after obtaining the network quality perception prediction result determined by the second target classifier according to the first classification result, the processor is further configured to implement:

[0089] Inputting the sample log data into a first preset classifier, and obtaining a sample perception prediction result of a second preset classifier based on the sample log data;

[0090] According to the preset network quality perception label and the sample perception prediction result, the parameters of the first preset classifier and the second preset classifier are adjusted until the first target classifier and the second target classifier are obtained; wherein, the first preset classifier includes: at least one of: a random forest classifier, an adaptive boosting classifier, and a support vector machine classifier; the second preset classifier includes: a logistic regression classifier.

[0091] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working process of the network quality perception prediction device described above can refer to the corresponding process in the aforementioned network quality perception prediction method embodiment, and will not be repeated here.

[0092] An embodiment of the present application also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any network quality perception prediction method provided in the embodiment of the present application.

[0093] The storage medium may be an internal storage unit of the network quality perception prediction device described in the aforementioned embodiment, such as a hard disk or memory of the network quality perception prediction device. The storage medium may also be an external storage device of the network quality perception prediction device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the network quality perception prediction device.

[0094] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable, and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic 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. In addition, it is well known to those skilled in the art that communication media generally embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0095] The serial numbers of the embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments. The above description is only a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A network quality perception prediction method, comprising: Obtain the user's target log data within a preset time period; Classifying the target log data according to the business type to which the target log data belongs, and obtaining target business data corresponding to at least one preset business tag; Inputting the target service data corresponding to each of the preset service tags into the first target classifier corresponding to each of the preset service tags to obtain a first classification result; Each of the first classification results is input into a second target classifier to obtain a network quality perception prediction result determined by the second target classifier according to the first classification results.

2. The network quality perception prediction method according to claim 1, wherein: The step of obtaining target log data of a user within a preset time period includes: Obtain the original traffic data in the bearer device; Parsing the original traffic data to obtain Internet access log data; The Internet access log data is filtered according to a preset filtering rule to obtain the target log data.

3. The network quality perception prediction method according to claim 2, wherein: The filtering of the online log data according to the preset filtering rules to obtain the target log data includes: The preset log data in the Internet access log data is filtered, and the preset log data includes: heartbeat signal data, small flow data with a flow value less than a preset value, and overseas flow data.

4. The network quality perception prediction method according to claim 1, wherein: The classifying the target log data according to the business type to which the target log data belongs, and obtaining the target business data corresponding to at least one preset business tag, includes: The target log data is clustered according to the service type field in the target log data to obtain the target service data corresponding to each of the preset service tags.

5. The network quality perception prediction method according to claim 1, wherein: The classifying the target log data according to the business type to which the target log data belongs, and obtaining the target business data corresponding to at least one preset business tag, includes: According to the preset service tags corresponding to the target log data, statistics are performed on the target service data of each preset service tag, and the target service data includes: at least one of uplink traffic data, downlink traffic data, duration data, log quantity data, delay data, and retransmission data.

6. The network quality perception prediction method according to claim 5, wherein: The step of inputting the target service data corresponding to each of the preset service tags into the first target classifier corresponding to each of the preset service tags to obtain a first classification result includes: In the case that the preset service tag does not have corresponding target service data, a preset value is input into the first target classifier as the target service data corresponding to the preset service tag.

7. The network quality perception prediction method according to claim 1, wherein: The step of inputting each of the first classification results into a second target classifier to obtain a network quality perception prediction result determined by the second target classifier according to the first classification result includes: splicing the first classification results to obtain a target input result; The target input result is input into the second target classifier to obtain the network quality perception prediction result, wherein the network quality perception prediction result includes: a positive perception result and a negative perception result.

8. The network quality perception prediction method according to any one of claims 1 to 7, wherein: The method further comprises: Input the sample log data into a first preset classifier, and obtain a sample perception prediction result of a second preset classifier based on the sample log data; According to the preset network quality perception label and the sample perception prediction result, the parameters of the first preset classifier and the second preset classifier are adjusted until the first target classifier and the second target classifier are obtained; wherein the first preset classifier includes: at least one of a random forest classifier, an adaptive boosting classifier, and a support vector machine classifier; the second preset classifier includes: a logistic regression classifier.

9. A network quality perception prediction device, comprising a processor, a memory, a computer program stored in the memory and executable by the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein when the computer program is executed by the processor, the steps of the network quality perception prediction method as described in any one of claims 1 to 8 are implemented.

10. A storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the network quality perception prediction method described in any one of claims 1 to 8.

Images