Method and device for determining poor network quality and electronic equipment

Abstract

The invention discloses a method and a device for determining poor network quality and electronic equipment. The method comprises the following steps: acquiring network flow data associated with Internet surfing perception experience, and determining a user multi-dimensional wide table according to the network flow data; according to index data in the user multi-dimensional wide table, a causal relationship logic diagram is determined, and the index data comprises network performance indexes, internet surfing perception indexes and factors affecting the network performance indexes or the internet surfing perception indexes; clustering the index data according to the causal relationship logic diagram to obtain a clustering result, and performing association analysis on the index data to obtain an association result; according to the clustering result and the association result, determining the comprehensive poor-quality confidence coefficient of the index data of each dimension; according to the comprehensive poor quality confidence coefficient, a poor quality delimited field is determined, and the poor quality delimited field is used for representing the dimensionality to which the index data belong when the index data have the quality difference. According to the invention, the technical problem of precision error existing in a mobile network networking experience poor quality delimiting method in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of network communications, and more specifically, to a method, device, and electronic device for determining poor network quality. Background Art

[0002] In related technologies, methods for determining the quality demarcation of users' mobile network Internet experience include a single-feature tag fixed threshold method, a multi-feature tag quality difference decision method, and a scenario-based multi-feature tag quality difference demarcation method. All of the above methods have precision errors and cannot find the ultimate cause of the problem.

[0003] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0004] The embodiments of the present application provide a method, device, and electronic device for determining poor network quality, so as to at least solve the technical problem of accuracy error in methods for defining poor mobile network Internet experience quality in related technologies.

[0005] According to one aspect of an embodiment of the present application, a method for determining poor network quality is provided, comprising: obtaining network traffic data associated with Internet access perception experience, and determining a user multi-dimensional wide table based on the network traffic data; determining a causal relationship logic diagram based on indicator data in the user multi-dimensional wide table, wherein the indicator data includes network performance indicators, Internet access perception indicators, and factors affecting network performance indicators or Internet access perception indicators; clustering the indicator data based on the causal relationship logic diagram to obtain clustering results, and performing correlation analysis on the indicator data to obtain correlation results; determining a comprehensive quality difference confidence level of the indicator data of each dimension based on the clustering results and the correlation results; determining a quality difference delimiting domain based on the comprehensive quality difference confidence level, wherein the quality difference delimiting domain is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

[0006] Optionally, a multi-dimensional wide table of users is determined based on network traffic data, including: extracting key network entities from network traffic data based on deep packet inspection technology, where each key network entity is a dimension; determining indicator data corresponding to the key network entities in the network traffic data; and determining a multi-dimensional wide table of users based on the key network entities and the indicator data.

[0007] Optionally, based on the indicator data in the user's multi-dimensional wide table, a causal relationship logic diagram is determined, including: obtaining the link indicators and link domains to which each feature data in the user's multi-dimensional wide table belongs, wherein the feature data includes key network entities and indicator data, the link indicators include uplink indicators, full link indicators and downlink indicators, and the link domain includes uplink delimiting dimensions and downlink delimiting dimensions; based on the link indicators and link domains, a causal relationship logic diagram is determined.

[0008] Optionally, the indicator data is clustered according to the causal relationship logic diagram to obtain a clustering result, including: determining the number of clusters according to the number of link indicators in the causal relationship logic diagram, and determining the initial cluster center; clustering the indicator data under each dimension according to the number of clusters, and updating the cluster center; until the cluster center no longer changes or the maximum number of iterations is reached, clustering is completed to obtain a clustering result.

[0009] Optionally, correlation analysis is performed on the indicator data to obtain correlation results, including: obtaining the first indicator data and the second indicator data under the target dimension, and obtaining the first indicator average value corresponding to the first indicator data and the second indicator average value corresponding to the second indicator data, wherein the target dimension is any key network entity in the user's multi-dimensional wide table, and the first indicator data and the second indicator data are any two indicator data under the target dimension; determining the first correlation result of the first indicator data and the second indicator data based on the first indicator data, the second indicator data, the first indicator average value and the second indicator average value; determining the correlation result based on the first correlation result of any two indicator data under all dimensions.

[0010] Optionally, based on the clustering results and the association results, the comprehensive quality difference confidence of the indicator data of each dimension is determined, including: obtaining the first indicator data, the second indicator data, the first indicator average value, the second indicator average value and the first association result; determining the first calibration confidence of the second indicator data for calibrating the first indicator data based on the first indicator data, the second indicator data, the first indicator average value, the second indicator average value and the first association result; determining the second association result of each other indicator data under the target dimension and the first indicator data, and determining the second calibration confidence of the other indicator data for calibrating the first indicator data based on the second association result, wherein the other indicator data is any indicator data under the target dimension except the first indicator data and the second indicator data; determining the comprehensive quality difference confidence of the first indicator data based on the first calibration confidence and the second calibration confidence.

[0011] Optionally, the quality difference delimiting domain is determined based on the comprehensive quality difference confidence, including: determining the comprehensive quality difference confidence corresponding to the indicator data of each dimension to obtain a comprehensive quality difference confidence set; sorting the comprehensive quality difference confidences in the comprehensive quality difference confidence set in order of confidence from high to low to obtain a sorted comprehensive quality difference confidence set; determining the dimensions corresponding to the first N confidences in the sorted comprehensive quality difference confidence set as the quality difference delimiting domain, where N is a positive integer.

[0012] According to another aspect of an embodiment of the present application, a device for determining network quality difference is also provided, including: an acquisition module, used to obtain network traffic data associated with the Internet perception experience, and determine a user multi-dimensional wide table based on the network traffic data; a first determination module, used to determine a causal relationship logic diagram based on the indicator data in the user multi-dimensional wide table, wherein the indicator data is used to include network performance indicators, Internet perception indicators and factors affecting network performance indicators or Internet perception indicators; an analysis module, used to cluster the indicator data according to the causal relationship logic diagram to obtain clustering results, and perform correlation analysis on the indicator data to obtain correlation results; a second determination module, used to determine the comprehensive quality difference confidence of the indicator data of each dimension based on the clustering results and the correlation results; a third determination module, used to determine the quality difference delimiting domain based on the comprehensive quality difference confidence, wherein the quality difference delimiting domain is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

[0013] According to another aspect of the embodiments of the present application, an electronic device is also provided, including: a memory for storing program instructions; a processor, connected to the memory, for executing program instructions to implement the following functions: obtaining network traffic data associated with Internet perception experience, and determining a user multi-dimensional wide table based on the network traffic data; determining a causal relationship logic diagram based on indicator data in the user multi-dimensional wide table, wherein the indicator data includes network performance indicators, Internet perception indicators and factors affecting network performance indicators or Internet perception indicators; clustering the indicator data according to the causal relationship logic diagram to obtain clustering results, and performing correlation analysis on the indicator data to obtain correlation results; determining the comprehensive quality difference confidence of the indicator data of each dimension based on the clustering results and the correlation results; determining the quality difference delimiting domain based on the comprehensive quality difference confidence, wherein the quality difference delimiting domain is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

[0014] According to another aspect of the embodiments of the present application, a non-volatile storage medium is provided, which includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the above-mentioned method for determining network quality difference by running the computer program.

[0015] According to another aspect of the embodiments of the present application, a computer program product is provided, including computer instructions, which implement the above-mentioned method for determining network quality difference when executed by a processor.

[0016] In an embodiment of the present application, network traffic data associated with the Internet perception experience is obtained, and a user multi-dimensional wide table is determined based on the network traffic data; a causal relationship logic diagram is determined based on the indicator data in the user multi-dimensional wide table, wherein the indicator data includes network performance indicators, Internet perception indicators and factors affecting network performance indicators or Internet perception indicators; the indicator data is clustered according to the causal relationship logic diagram to obtain clustering results, and the indicator data is subjected to correlation analysis to obtain correlation results; based on the clustering results and the correlation results, the comprehensive quality difference confidence of the indicator data of each dimension is determined; based on the comprehensive quality difference confidence, a quality difference demarcation domain is determined, wherein the quality difference demarcation domain is used to indicate the dimension to which the indicator data belongs when there is a quality difference, thereby achieving the purpose of accurately locating the network quality difference problem domain, thereby achieving the technical effect of improving the accuracy of the quality difference demarcation of the mobile network Internet experience, and further solving the technical problem of the accuracy error in the quality difference demarcation method of the mobile network Internet experience in the related technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 is a hardware structure block diagram of a computer terminal for implementing a method for determining poor network quality according to an embodiment of the present application;

[0019] Figure 2 is a flow chart of a method for determining poor network quality according to an embodiment of the present application;

[0020] Figure 3 is a schematic diagram of a user multi-dimensional wide table according to an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of a cause-effect relationship logic diagram according to an embodiment of the present application;

[0022] Figure 5 is a schematic diagram of an association result according to an embodiment of the present application;

[0023] Figure 6 This is a structural diagram of a device for determining poor network quality according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] The information collected in the embodiments of the present application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or reject the automated decision results; if the user chooses to reject, the expert decision-making process will be entered.

[0027] First, some nouns or terms that appear in the process of explaining the embodiments of this application are subject to the following explanations:

[0028] DPI (Deep Packet Inspection) is a network security and management technology used to analyze, filter, and monitor data packets on a network. Unlike traditional network monitoring technologies that only examine packet header information (such as source address, destination address, and port number), DPI technology delves into the packet's payload, the internal data, for detailed analysis.

[0029] TCP (Transmission Control Protocol): A connection-oriented, reliable, byte-stream-based transport layer communications protocol. TCP plays a core role in the transport layer of the OSI model and the transport layer of the TCP / IP model, providing reliable, end-to-end data transmission services over unreliable internetworks.

[0030] UPF (User Plane Function): The network element on the 5G core network side is responsible for the transmission and forwarding of user plane data.

[0031] S / PGW: A converged network element on the 4G core network side responsible for the transmission, forwarding, and control of user-plane data, including SGW (Serving Gateway) and PGW (PDN Gateway).

[0032] Tracking Area Code (TAC): A code used in mobile communication networks to manage the location and location of mobile devices. To efficiently manage the location and paging of mobile terminals, mobile communication networks divide the coverage area into multiple Tracking Areas (TAs). Each TA is assigned a unique Tracking Area Code (TA).

[0033] In related technologies, the mainstream method for defining poor mobile network experience quality is based on the following logic:

[0034] 1. Single feature label fixed threshold method: This method uses a fixed threshold to determine poor-quality cells and users (for example, TCP uplink establishment success rate <= 80%). It does not consider the service feature logic and complexity, and directly outputs poor-quality results, resulting in accuracy errors.

[0035] 2. The multi-feature label quality decision method adds more feature indicators (for example, TCP uplink establishment success rate <= 80% and TCP uplink establishment delay > 70ms) on the basis of a single-feature fixed threshold, associates features for delimitation, and improves accuracy compared to a single feature.

[0036] 3. Quality difference delimitation of multi-feature labels based on scenes. Based on the multi-feature quality difference decision method, the features are constructed by considering factors such as scenes, and the matching output is used to determine the delimitation results.

[0037] The above-mentioned demarcation of poor quality of existing users' mobile network Internet experience has the following deficiencies and defects: the end-to-end demarcation logic of the Internet experience perception service is based on single-dimensional features and expert judgment, and has accuracy errors; the threshold for poor Internet experience perception quality is a fixed value or a relatively fixed value, which does not take into account multi-dimensional quality differences. The relatively static threshold cannot reflect the quality difference domain; the demarcation of poor Internet experience perception quality may contain multiple quality difference label features, and the demarcation results are ambiguous, making it impossible to accurately find the ultimate cause of the problem.

[0038] In order to solve the problems existing in the related art, the embodiment of the present application provides a method for determining poor network quality, which can be run on Figure 1 In the computer terminal shown, the computer terminal is explained below.

[0039] The method for determining poor network quality provided in the embodiments of the present application may be executed in a mobile terminal, a computer terminal, or a similar computing device. Figure 1 FIG1 shows a hardware structure block diagram of a computer terminal for implementing a method for determining poor network quality. Figure 1 As shown, the computer terminal 10 may include one or more (illustrated by 102a, 102b, ..., 102n in the figure) processors (the processor may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission module 106 for communication functions connected via a wired and / or wireless network. In addition, it may also include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0040] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry." The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be incorporated in whole or in part into any of the other components of the computer terminal 10. As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0041] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for determining poor network quality in the embodiment of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby implementing the above-mentioned method for determining poor network quality. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely located relative to the processor, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0042] The transmission module 106 is configured to receive or transmit data via a network. A specific example of the aforementioned network may include a wireless network provided by the communications provider of the computer terminal 10. In one embodiment, the transmission module 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission module 106 may be a radio frequency (RF) module, which is configured to communicate with the Internet wirelessly.

[0043] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0044] It should be noted that, in some optional embodiments, the above Figure 1 The computer terminal shown may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of hardware elements and software elements. Figure 1 This is merely one example of a particular embodiment and is intended to illustrate the types of components that may be present in the computer terminal described above.

[0045] In the above-mentioned operating environment, an embodiment of the present application provides an embodiment of a method for determining poor network quality. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0046] Figure 2 is a flow chart of a method for determining poor network quality according to an embodiment of the present application, such as Figure 2As shown, the method includes the following steps:

[0047] Step S202: obtaining network traffic data associated with Internet access perception experience, and determining a user multi-dimensional wide table based on the network traffic data.

[0048] In the aforementioned step S202, in the field of mobile network communications, the Internet experience refers to the actual performance felt by the terminal user when using network services, such as data transmission rate, latency, connectivity, etc. Network traffic data includes, but is not limited to: user behavior data (including websites, applications, service request duration, data volume, etc. visited by the user), network performance data (such as TCP establishment success rate, TCP latency, uplink and downlink data transmission rate, packet loss rate, retransmission rate, radio signal strength (RSRP), uplink and downlink RTT (round trip time), etc.), terminal device information (including terminal model, operating system version, software application version, etc.), network entity information (such as base station, cell (TAC), gateway (S / PGW, UPF) information), service type data, etc.

[0049] Before building a user multi-dimensional wide table, it is necessary to clean and pre-process the acquired network traffic data, including cleaning the acquired raw network traffic data and removing invalid or incomplete records to ensure the quality and integrity of the data. Based on the pre-processed network traffic data, build a user multi-dimensional wide table, such as Figure 3 shown.

[0050] Step S204 , determining a cause-and-effect relationship logic diagram based on the indicator data in the user multi-dimensional wide table, wherein the indicator data includes network performance indicators, Internet access perception indicators, and factors affecting the network performance indicators or Internet access perception indicators.

[0051] In the above step S204, the indicator data of the user's multi-dimensional wide table is obtained, such as TCP establishment success rate, TCP delay, video download rate, game delay, etc., and the link indicators and link domains corresponding to different indicator data are determined. Based on all the link indicators and link domains, the causal relationship logic diagram is determined.

[0052] Step S206: cluster the indicator data according to the cause-effect relationship logic diagram to obtain clustering results, and perform correlation analysis on the indicator data to obtain correlation results.

[0053] In the above step S206 , the indicator data are clustered according to the link indicators in the causal relationship logic diagram to obtain clustering results, and the associated indicators of each indicator data are determined to obtain the associated results of each indicator data.

[0054] Step S208: Determine the comprehensive quality confidence of the indicator data of each dimension based on the clustering results and the association results.

[0055] In the above step S208, the comprehensive quality difference confidence is used to evaluate the quality difference possibility of the indicator data of each dimension in a quantitative manner.

[0056] In step S210 , a quality difference delimiting region is determined based on the comprehensive quality difference confidence level, wherein the quality difference delimiting region is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

[0057] In step S210, the determination of poor quality boundaries is based on a multi-dimensional, comprehensive poor quality confidence analysis of the user's online experience, used to accurately locate the source of network issues that degrade the user experience. A poor quality boundary refers to a dimension across multiple network dimensions (such as time, location, device type, and service type) where the indicator data in that dimension exhibits significantly lower quality than the expected or standard level. This dimension is identified as a poor quality boundary.

[0058] Through steps S202 to S210, the purpose of accurately locating the problem domain of poor network quality is achieved, thereby achieving the technical effect of improving the accuracy of demarcating poor mobile network experience quality, and further solving the technical problem of accuracy error in the demarcation method of poor mobile network experience quality in related technologies. The following is an explanation.

[0059] In step S202 of the above-mentioned method for determining poor network quality, a multi-dimensional wide table of users is determined based on network traffic data, including: extracting key network entities from the network traffic data based on deep packet inspection technology, wherein each key network entity is a dimension; determining indicator data corresponding to the key network entities in the network traffic data; and determining a multi-dimensional wide table of users based on the key network entities and the indicator data.

[0060] In some embodiments of the present application, DPI is a network packet analysis method that can parse network traffic at the packet level and extract detailed information about specific applications, protocols, devices, or network components. Through DPI technology, the following key network entities can be identified and extracted from network traffic data, including: service type, network element information, location information (including TAV, base station, cell), terminal, etc. Each key network entity constitutes a dimension of the user multi-dimensional wide table, representing a different level or perspective of analysis, and extracts indicator data related to the key network entity from the network traffic data, such as: uplink and downlink TCP establishment success rate and delay, video download rate, game delay, signal strength (RSRP), packet loss rate, and retransmission rate. Based on the extracted key network entities and their corresponding indicator data, a user multi-dimensional wide table is constructed, wherein each column in the user multi-dimensional wide table corresponds to a key network entity or indicator data, such as a terminal model column, a TCP uplink establishment success rate column, etc. Each row represents an analysis unit such as a specific time point, user, or session, and contains indicator data related to all dimensions.

[0061] In step S204 of the above-mentioned method for determining poor network quality, a causal relationship logic diagram is determined based on the indicator data in the user's multi-dimensional wide table, including: obtaining the link indicators and link domains to which each feature data in the user's multi-dimensional wide table belongs, wherein the feature data includes key network entities and indicator data, the link indicators include uplink indicators, full link indicators and downlink indicators, and the link domain includes uplink delimiting dimensions and downlink delimiting dimensions; and determining the causal relationship logic diagram based on the link indicators and link domains.

[0062] In some embodiments of the present application, based on the link classification theory, the network performance indicators are mapped to their positions in the network architecture, including identifying and classifying different feature data from the user's multi-dimensional wide table, and associating them with link indicators. For example, uplink indicators: related to the performance of the user terminal uploading data to the network, such as TCP uplink establishment success rate, uplink RTT delay, etc.; full-link indicators: end-to-end performance indicators from the user terminal to the server, covering uplink and downlink links, such as video download rate, game delay, etc.; downlink indicators: focus on the quality of data transmitted by the network to the user terminal, such as downlink TCP establishment success rate, downlink RTT delay excellence rate, etc. The above link indicators are further mapped to the network entities they may affect, namely the uplink delimitation dimension and the downlink delimitation dimension. Each dimension represents a potential impact range in network operation and maintenance, for example: uplink delimitation dimension: including entities such as terminal model, terminal software version, base station, cell, etc.; downlink delimitation dimension: involving cities, districts and counties, location tracking areas (TACs), network element types such as UPF or SBC, etc. Based on the link indicators and link domains, a causal logic diagram is determined, such as Figure 4 shown.

[0063] In step S206 of the above-mentioned method for determining poor network quality, the indicator data is clustered according to the causal relationship logic diagram to obtain a clustering result, including: determining the number of clusters according to the number of link indicators in the causal relationship logic diagram, and determining the initial cluster center; clustering the indicator data under each dimension according to the number of clusters, and updating the cluster center; until the cluster center no longer changes or the maximum number of iterations is reached, clustering is completed to obtain a clustering result.

[0064] In some embodiments of the present application, for example, the K-means algorithm is selected to implement clustering. Specifically, the preliminary number of clusters is determined based on the number of link indicators in the causal relationship logic diagram. For example, when the link indicator includes an uplink indicator, a full link indicator, and a downlink indicator, the corresponding number of clusters is 3, and the initial cluster center (for example, 3) is randomly determined. The indicator data points under each dimension are assigned to the nearest cluster center to form a preliminary cluster group, wherein the distance metric is usually determined based on the Euclidean distance or the Manhattan distance. The average or median of all indicator data points in each cluster is calculated to update the position of the cluster center. Repeat the data allocation (i.e., the above-mentioned process of assigning the indicator data points under each dimension to the nearest cluster center) and the process of updating the cluster center until the cluster center no longer changes significantly or reaches the preset maximum number of iterations, then the clustering is completed and the clustering result is obtained. Furthermore, internal clustering indicators (such as the average distance within the cluster) and external clustering indicators (such as comparison with other known categories) can be used to evaluate the clustering quality. If the quality is poor, you may need to adjust the number of clusters or reinitialize the cluster centers to optimize the clustering results. Based on the clustering effect and network operation and maintenance requirements, fine-tune the algorithm parameters, such as the maximum number of iterations and distance measurement method, to achieve the best clustering effect.

[0065] When performing clustering, automatic clustering can be performed at least 7 days * 24 hours to avoid result fluctuations caused by normal network characteristics such as network busyness and idleness to a certain extent.

[0066] In step S206 of the above-mentioned method for determining poor network quality, correlation analysis is performed on the indicator data to obtain correlation results, including: obtaining first indicator data and second indicator data under the target dimension, and obtaining a first indicator average value corresponding to the first indicator data and a second indicator average value corresponding to the second indicator data, wherein the target dimension is any key network entity in the user's multi-dimensional wide table, and the first indicator data and the second indicator data are any two indicator data under the target dimension; determining a first correlation result of the first indicator data and the second indicator data based on the first indicator data, the second indicator data, the first indicator average value and the second indicator average value; determining a correlation result based on the first correlation result of any two indicator data under all dimensions.

[0067] In some embodiments of the present application, the first correlation result is determined by the following formula:

[0068]

[0069] Among them, r represents the first correlation result, x i Represents the first indicator data under the target dimension, y i Indicates the second indicator data under the target dimension. represents the average value of the first indicator, Represents the average value of the second indicator, i represents the amount of observed data, which refers to the number of online users under each dimension. The set of first correlation results corresponding to all dimensions is determined as the above correlation result, wherein data greater than or equal to 40% of the correlation results are retained to obtain the Top-n correlation results, such as Figure 5 shown.

[0070] In step S208 of the above-mentioned method for determining network quality difference, the comprehensive quality difference confidence of the indicator data of each dimension is determined based on the clustering results and the association results, including: obtaining the first indicator data, the second indicator data, the first indicator average value, the second indicator average value and the first association result; determining the first calibration confidence of the second indicator data for calibrating the first indicator data based on the first indicator data, the second indicator data, the first indicator average value, the second indicator average value and the first association result; determining the second association result of each other indicator data under the target dimension with the first indicator data, and determining the second calibration confidence of the other indicator data for calibrating the first indicator data based on the second association result, wherein the other indicator data is any indicator data under the target dimension except the first indicator data and the second indicator data; determining the comprehensive quality difference confidence of the first indicator data based on the first calibration confidence and the second calibration confidence.

[0071] In some embodiments of the present application, the first calibration confidence is determined by the following formula:

[0072]

[0073] Among them, Z j The second indicator data is the first indicator data x i The first calibration confidence, r y is x i and y i The Pearson correlation of the indicator (or the first indicator data x i and the second indicator data y i The first correlation result between y and N represents the total amount of observation data. i Indicator pair x i The impact of indicators for xi In the calibration and correction of indicators, long-term, short-term, and emergency scenarios are evaluated and handled to obtain high-confidence x i Verification indicators.

[0074] The formula corresponding to the comprehensive quality difference confidence is as follows:

[0075]

[0076] Among them, z i Represents the first indicator data x i The corrected confidence level (the confidence level corrected according to the first calibration confidence level and the second calibration confidence level), Z represents the comprehensive quality difference confidence level of the first indicator data.

[0077] In step S210 of the above-mentioned method for determining network quality difference, the quality difference delimiting domain is determined based on the comprehensive quality difference confidence, including: determining the comprehensive quality difference confidence corresponding to the indicator data of each dimension to obtain a comprehensive quality difference confidence set; sorting the comprehensive quality difference confidences in the comprehensive quality difference confidence set in order of confidence from high to low to obtain a sorted comprehensive quality difference confidence set; determining the dimensions corresponding to the first N confidences in the sorted comprehensive quality difference confidence set as the quality difference delimiting domain, where N is a positive integer.

[0078] In some embodiments of the present application, the comprehensive quality difference confidence corresponding to the indicator data of each dimension reflects whether the degree of network performance deviation or decline perceived by users in a specific dimension of the network is credible. For example, the performance of a base station in a specific time period, or the video download speed on a certain model of mobile phone, will have a corresponding comprehensive quality difference confidence value, and all comprehensive quality difference confidence values constitute a comprehensive quality difference confidence set. The comprehensive quality difference confidence set is sorted in order from high to low according to the confidence value. The sorted comprehensive quality difference confidence set can intuitively show which dimensions of network performance problems are most likely to exist, and the importance level of these problems. According to the operation and maintenance strategy and resource constraints, a positive integer N is selected to represent the number of quality difference delimiters that you want to investigate and process first. In the sorted comprehensive quality difference confidence set, the dimensions corresponding to the top N highest confidence values are selected as the quality difference delimiters.

[0079] Figure 6 is a structural diagram of a device for determining poor network quality according to an embodiment of the present application, such as Figure 6 As shown, the device includes:

[0080] An acquisition module 40 is configured to acquire network traffic data associated with Internet access perception experience and determine a user multi-dimensional wide table based on the network traffic data;

[0081] A first determination module 42 is configured to determine a cause-and-effect relationship logic diagram based on indicator data in the user multi-dimensional wide table, wherein the indicator data includes network performance indicators, Internet access perception indicators, and factors affecting the network performance indicators or Internet access perception indicators;

[0082] The analysis module 44 is used to cluster the indicator data according to the cause-effect relationship logic diagram to obtain clustering results, and to perform correlation analysis on the indicator data to obtain correlation results;

[0083] The second determination module 46 is used to determine the comprehensive quality confidence of the indicator data of each dimension based on the clustering results and the correlation results;

[0084] The third determining module 48 is used to determine a quality difference delimiting region based on the comprehensive quality difference confidence, wherein the quality difference delimiting region is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

[0085] Through the acquisition module, first determination module, analysis module, second determination module and third determination module in the above-mentioned network quality determination device, the purpose of accurately locating the network quality problem domain is achieved, thereby achieving the technical effect of improving the accuracy of demarcation of mobile network Internet experience quality differences, and further solving the technical problem of accuracy error in the mobile network Internet experience quality difference demarcation method in related technologies.

[0086] In the acquisition module in the above-mentioned network quality determination device, the acquisition module is also used to extract key network entities from network traffic data based on deep packet inspection technology, where each key network entity is a dimension; in the network traffic data, determine the indicator data corresponding to the key network entity; based on the key network entity and the indicator data, determine the user multi-dimensional wide table.

[0087] In the first determination module in the above-mentioned network quality determination device, the first determination module is also used to obtain the link indicators and link domains to which each feature data in the user's multi-dimensional wide table belongs, wherein the feature data includes key network entities and indicator data, the link indicators include uplink indicators, full link indicators and downlink indicators, and the link domain includes uplink delimiting dimensions and downlink delimiting dimensions; based on the link indicators and link domains, a causal relationship logic diagram is determined.

[0088] In the analysis module in the above-mentioned network quality determination device, the analysis module is also used to determine the number of clusters based on the number of link indicators in the causal logic diagram, and to determine the initial cluster center; cluster the indicator data under each dimension according to the number of clusters, and update the cluster center; until the cluster center no longer changes or the maximum number of iterations is reached, clustering is completed to obtain the clustering result.

[0089] In the analysis module in the above-mentioned network quality determination device, the analysis module is also used to obtain the first indicator data and the second indicator data under the target dimension, and obtain the first indicator average value corresponding to the first indicator data and the second indicator average value corresponding to the second indicator data, wherein the target dimension is any key network entity in the user's multi-dimensional wide table, and the first indicator data and the second indicator data are any two indicator data under the target dimension; based on the first indicator data, the second indicator data, the first indicator average value and the second indicator average value, determine the first correlation result of the first indicator data and the second indicator data; based on the first correlation result of any two indicator data under all dimensions, determine the correlation result.

[0090] In the second determination module in the above-mentioned network quality difference determination device, the second determination module is also used to obtain the first indicator data, the second indicator data, the first indicator average value, the second indicator average value and the first correlation result; determine the first calibration confidence of the second indicator data for calibrating the first indicator data based on the first indicator data, the second indicator data, the first indicator average value, the second indicator average value and the first correlation result; determine the second correlation result of each other indicator data under the target dimension and the first indicator data, and determine the second calibration confidence of the other indicator data for calibrating the first indicator data based on the second correlation result, wherein the other indicator data is any indicator data under the target dimension except the first indicator data and the second indicator data; determine the comprehensive quality difference confidence of the first indicator data based on the first calibration confidence and the second calibration confidence.

[0091] In the third determination module in the above-mentioned network quality difference determination device, the third determination module is also used to determine the comprehensive quality difference confidence corresponding to the indicator data of each dimension to obtain a comprehensive quality difference confidence set; sort the comprehensive quality difference confidences in the comprehensive quality difference confidence set in order from high to low confidence to obtain a sorted comprehensive quality difference confidence set; determine the dimensions corresponding to the first N confidences in the sorted comprehensive quality difference confidence set as the quality difference delimiter, where N is a positive integer.

[0092] It should be noted that Figure 6 The network quality determination device shown is used to perform Figure 2 The method for determining the network quality difference shown in FIG. 4 is used. Therefore, the relevant explanations in the above method for determining the network quality difference are also applicable to the device for determining the network quality difference, and will not be repeated here.

[0093] An embodiment of the present application also provides an electronic device, which includes a memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute program instructions to implement the following functions: obtaining network traffic data associated with Internet perception experience, and determining a user multi-dimensional wide table based on the network traffic data; determining a causal relationship logic diagram based on indicator data in the user multi-dimensional wide table, wherein the indicator data includes network performance indicators, Internet perception indicators and factors affecting network performance indicators or Internet perception indicators; clustering the indicator data based on the causal relationship logic diagram to obtain clustering results, and performing correlation analysis on the indicator data to obtain correlation results; determining the comprehensive quality difference confidence of the indicator data of each dimension based on the clustering results and the correlation results; determining the quality difference delimiting domain based on the comprehensive quality difference confidence, wherein the quality difference delimiting domain is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

[0094] It should be noted that the above electronic equipment is used to perform Figure 2 The method for determining poor network quality is shown in FIG. 4 , so the relevant explanations in the above method for determining poor network quality are also applicable to the electronic device and will not be repeated here.

[0095] An embodiment of the present application also provides a non-volatile storage medium, which includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the following method for determining network quality difference by running the computer program: obtaining network traffic data associated with Internet access perception experience, and determining a user multi-dimensional wide table based on the network traffic data; determining a causal relationship logic diagram based on indicator data in the user multi-dimensional wide table, wherein the indicator data includes network performance indicators, Internet access perception indicators, and factors affecting network performance indicators or Internet access perception indicators; clustering the indicator data based on the causal relationship logic diagram to obtain clustering results, and performing correlation analysis on the indicator data to obtain correlation results; determining the comprehensive quality difference confidence of the indicator data of each dimension based on the clustering results and the correlation results; determining a quality difference delimiting domain based on the comprehensive quality difference confidence, wherein the quality difference delimiting domain is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

[0096] It should be noted that the above non-volatile storage medium is used to execute Figure 2 The method for determining the network quality difference is shown in FIG. 4 , so the relevant explanations in the above method for determining the network quality difference are also applicable to the non-volatile storage medium and will not be repeated here.

[0097] An embodiment of the present application further provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the method for determining poor network quality in each embodiment of the present application.

[0098] The embodiments of the present application further provide a computer program, which, when executed by a processor, implements the steps of the method for determining poor network quality in each embodiment of the present application.

[0099] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0100] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0102] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0103] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0104] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0105] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining poor network quality, characterized in that: include: Obtaining network traffic data associated with online experience, and determining a user multi-dimensional wide table based on the network traffic data; Determining a causal relationship logic diagram based on the indicator data in the user multi-dimensional wide table, wherein the indicator data includes a network performance indicator, an Internet access perception indicator, and factors affecting the network performance indicator or the Internet access perception indicator; Clustering the indicator data according to the causal relationship logic diagram to obtain a clustering result, and performing association analysis on the indicator data to obtain an association result; Determining the comprehensive quality confidence of the indicator data of each dimension based on the clustering results and the correlation results; A quality difference delimiting domain is determined based on the comprehensive quality difference confidence, wherein the quality difference delimiting domain is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

2. The method according to claim 1, characterized in that Determining a user multi-dimensional wide table based on the network traffic data includes: Extracting key network entities from the network traffic data based on deep packet inspection technology, wherein each key network entity is a dimension; Determining, from the network traffic data, indicator data corresponding to the key network entity; The user multi-dimensional wide table is determined based on the key network entity and the indicator data.

3. The method according to claim 2, characterized in that Based on the indicator data in the user multi-dimensional wide table, a causal relationship logic diagram is determined, including: Obtaining a link indicator and a link domain to which each feature data in the user multi-dimensional wide table belongs, wherein the feature data includes the key network entity and the indicator data, the link indicator includes an uplink indicator, a full link indicator, and a downlink indicator, and the link domain includes an uplink delimiting dimension and a downlink delimiting dimension; The causal relationship logic diagram is determined based on the link indicator and the link domain.

4. The method according to claim 3, characterized in that Clustering the indicator data according to the causal relationship logic diagram to obtain clustering results includes: Determining the number of clusters according to the number of link indicators in the causal relationship logic diagram and determining the initial cluster centers; Clustering the indicator data under each dimension according to the number of clusters, and updating the cluster center; The clustering is completed until the cluster center no longer changes or the maximum number of iterations is reached, and the clustering result is obtained.

5. The method according to claim 2, characterized in that Perform correlation analysis on the indicator data to obtain correlation results, including: Obtaining first indicator data and second indicator data under a target dimension, and obtaining a first indicator average value corresponding to the first indicator data and a second indicator average value corresponding to the second indicator data, wherein the target dimension is any key network entity in the user multi-dimensional wide table, and the first indicator data and the second indicator data are any two indicator data under the target dimension; Determining a first correlation result between the first indicator data and the second indicator data based on the first indicator data, the second indicator data, the first indicator average value, and the second indicator average value; The correlation result is determined based on the first correlation result of any two indicator data in all dimensions.

6. The method according to claim 5, characterized in that Determine the comprehensive quality confidence of the indicator data of each dimension based on the clustering results and the correlation results, including: Obtaining the first indicator data, the second indicator data, the first indicator average value, the second indicator average value, and the first correlation result; Determining a first calibration confidence level for calibrating the first indicator data with the second indicator data based on the first indicator data, the second indicator data, the first indicator average value, the second indicator average value, and the first correlation result; Determine a second correlation result between each other indicator data under the target dimension and the first indicator data, and determine a second calibration confidence level for calibrating the first indicator data using the other indicator data based on the second correlation result, wherein the other indicator data is any indicator data under the target dimension other than the first indicator data and the second indicator data; The comprehensive quality difference confidence of the first indicator data is determined based on the first calibration confidence and the second calibration confidence.

7. The method according to claim 1, characterized in that Determining a quality difference boundary region based on the comprehensive quality difference confidence level includes: Determine the comprehensive quality difference confidence level corresponding to the indicator data of each dimension and obtain a comprehensive quality difference confidence level set; Sorting the comprehensive quality difference confidences in the comprehensive quality difference confidence set in descending order of confidence to obtain a sorted comprehensive quality difference confidence set; The dimensions corresponding to the first N confidences in the sorted comprehensive quality difference confidence set are determined as the quality difference delimiting domain, where N is a positive integer.

8. A device for determining poor network quality, characterized in that: include: An acquisition module, configured to acquire network traffic data associated with Internet access perception experience and determine a user multi-dimensional wide table based on the network traffic data; A first determination module is configured to determine a causal relationship logic diagram based on indicator data in the user multi-dimensional wide table, wherein the indicator data is configured to include a network performance indicator, an Internet access perception indicator, and factors affecting the network performance indicator or the Internet access perception indicator; An analysis module, configured to cluster the indicator data according to the causal relationship logic diagram to obtain a clustering result, and perform association analysis on the indicator data to obtain an association result; A second determination module is used to determine the comprehensive quality difference confidence of the indicator data of each dimension based on the clustering result and the correlation result; The third determination module is used to determine a quality difference delimiting domain based on the comprehensive quality difference confidence, wherein the quality difference delimiting domain is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

9. An electronic device, characterized in that: include: a memory for storing program instructions; A processor, connected to the memory, is used to execute program instructions that implement the following functions: obtain network traffic data associated with Internet access perception experience, and determine a user multi-dimensional wide table based on the network traffic data; determine a causal relationship logic diagram based on indicator data in the user multi-dimensional wide table, wherein the indicator data includes network performance indicators, Internet access perception indicators and factors affecting the network performance indicators or the Internet access perception indicators; cluster the indicator data based on the causal relationship logic diagram to obtain clustering results, and perform correlation analysis on the indicator data to obtain correlation results; determine the comprehensive quality difference confidence of the indicator data of each dimension based on the clustering results and the correlation results; determine the quality difference demarcation domain based on the comprehensive quality difference confidence, wherein the quality difference demarcation domain is used to indicate the dimension to which the indicator data belongs when there is a quality difference.

10. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the method for determining network quality poor according to any one of claims 1 to 7 by running the computer program.

11. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the method for determining network quality poorness according to any one of claims 1 to 7 is implemented.

Images