Grid cell-based broadband quality dynamic evaluation method and device

Abstract

The application discloses a kind of broadband quality dynamic evaluation method and device based on grid unit.Therein, the method includes: obtaining multiple groups of first target data of multiple grid units in target area, wherein each group of first target data includes: at least one kind of first index data for reflecting the quality of network equipment in grid unit and at least one kind of second index data for reflecting the service evaluation of user to broadband service in grid unit;Determine the influence factor affecting the first target data, and modify multiple groups of first target data according to the influence factor respectively, obtain multiple groups of second target data;Determine the target weight of each kind of index data in multiple groups of second target data, and determine the broadband quality score of each grid unit based on target weight.The application solves the technical problem that related technology does not consider multi-dimensional multi-level index data, resulting in poor accuracy of broadband quality score.

Description

Technical Field

[0001] This application relates to the field of big data technology, and more specifically, to a method and apparatus for dynamic evaluation of broadband quality based on grid cells. Background Technology

[0002] Typically, network broadband quality is evaluated based on network KPIs (Key Performance Indicators), with a broadband quality score calculated by weighting simple indicators. However, this method often lacks multi-level and multi-dimensional indicator data that considers factors such as region, grid, and user. Therefore, single indicator data is insufficient to depict the user's real experience and cannot accurately reflect network broadband quality issues.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides a method and apparatus for dynamic evaluation of broadband quality based on grid cells, which at least solves the technical problem that related technologies do not consider multi-dimensional and multi-level indicator data, resulting in poor accuracy of broadband quality scoring.

[0005] According to one aspect of the embodiments of this application, a method for dynamic evaluation of broadband quality based on grid cells is provided, comprising: acquiring multiple sets of first target data for multiple grid cells within a target area, wherein each set of first target data includes: at least one type of first indicator data reflecting the quality of network devices within the grid cell and at least one type of second indicator data reflecting the service evaluation of broadband services by users within the grid cell; determining the influencing factors affecting the first target data, and correcting the multiple sets of first target data according to the influencing factors to obtain multiple sets of second target data; determining the target weight of each type of indicator data in the multiple sets of second target data, and determining the broadband quality score of each grid cell based on the target weight.

[0006] Optionally, multiple sets of first target data from multiple grid cells within the target area are acquired, including: for each grid cell, acquiring network device index data and user evaluation index data within the grid cell; preprocessing the network device index data to obtain at least one type of first index data, wherein the type of the first index data includes at least one of the following: number of network device failures, month-on-month comparison of network device failures, duration of network device failures, number of users affected by network device failures, and user quality degradation trend; preprocessing the user evaluation index data to obtain at least one type of second index data, wherein the type of the second index data includes at least one of the following: total number of users who complained and total number of users who filed complaints.

[0007] Optionally, the influencing factors include: a basic influence factor, a sensitivity factor, and a continuous deterioration factor. Determining the influencing factors affecting the first target data includes: for each grid cell, determining the basic influence factor corresponding to the grid cell based on the first target data, wherein the basic influence factor is obtained by multiplying a network element influence factor (reflecting the impact of different types of network elements within the grid cell on users) and a time influence factor (reflecting the impact of different fault occurrence times within the grid cell on users); determining the sensitivity factor corresponding to the grid cell based on the first target data, wherein the sensitivity factor includes: a complaint sensitivity factor (reflecting the impact of the number of user complaints within the grid cell on service evaluation), a service sensitivity factor (reflecting the sensitivity of different broadband services within the grid cell to the network impact), and a repeating fault sensitivity factor (reflecting the impact of the number of identical network faults within the grid cell on service evaluation); and determining the continuous deterioration factor corresponding to the grid cell based on the first target data, wherein the continuous deterioration factor reflects the impact of the network fault persistence status within the grid cell on service evaluation.

[0008] Optionally, the basic factors of influence corresponding to the grid unit are determined based on the first target data corresponding to the grid unit, including: for each grid unit, determining the grid element influence factors corresponding to different types of grid elements within the grid unit based on the preset grid element level mapping relationship; and determining the time influence factors corresponding to different fault occurrence times within the grid unit based on the preset fault time mapping relationship.

[0009] Optionally, the sensitivity factors corresponding to the grid units are determined based on the first target data corresponding to the grid units, including: for each grid unit, determining a first weight coefficient for the total number of complaining users, and calculating a complaint sensitivity factor based on the total number of reporting users, the total number of complaining users, and the first weight coefficient; determining the number of reporting users corresponding to different types of users and the number of users affected by the fault corresponding to different types of users, and determining a second influence coefficient for the number of reporting users and a third influence coefficient for the number of users affected by the fault, and calculating a business sensitivity factor based on the number of reporting users, the number of users affected by the fault, the second influence coefficient, and the third influence coefficient, wherein at least pre-defined high-value users are included among the different types of users; determining the number of users affected by repeated faults, the total number of users affected by faults within the target evaluation period, the number of users who repeatedly report faults, and the total number of reporting users within the target evaluation period, and calculating a repeated obstacle sensitivity factor based on the number of users affected by repeated faults, the total number of users affected by faults, the number of users who repeatedly report faults, and the total number of reporting users.

[0010] Optionally, the continuous deterioration factor corresponding to the grid cell is determined based on the first target data corresponding to the grid cell, including: for each grid cell, determining the number of network failure duration days, the number of evaluation days in the target month, the number of first network failures in the second half of the target evaluation period, the number of second network failures in the first half of the target evaluation period, the total number of network failures, the number of third network failures and the number of first user quality defects as of the target date, the number of fourth network failures and the number of second user quality defects on the day before the target date, and calculating the continuous deterioration factor based on the number of network failure duration days, the number of evaluation days in the target month, the number of first network failures, the number of second network failures, the total number of network failures, the number of third network failures and the number of first user quality defects, as well as the number of fourth network failures and the number of second user quality defects.

[0011] Optionally, each set of second target data includes: at least one type of third indicator data and at least one type of fourth indicator data. Multiple sets of first target data are modified according to the impact factors to obtain multiple sets of second target data, including: for each set of first target data, modifying each type of first indicator data in the first target data based on the impact factor to obtain at least one type of third indicator data; modifying each type of second indicator data in the first target data based on the impact factor, the sensitivity factor, and the continuous deterioration factor to obtain at least one type of fourth indicator data.

[0012] Optionally, before determining the target weight of each type of indicator data in the multiple sets of second target data, the method further includes: performing positiveization and normalization processing on each type of indicator data in the multiple sets of second target data.

[0013] Optionally, determining the target weight for each type of indicator data in multiple sets of second target data includes: for each grid cell, using the entropy method to determine the initial weight of each type of indicator data in the second target data corresponding to the grid cell; using a variable weight function to determine the weight influence factor of the initial weight of each type of indicator data in the second target data, wherein the weight influence factor is used to reflect the influence of the degree of change and duration of change of each type of indicator data in the second target data on the initial weight within the target time period; and calculating the target weight of each type of indicator data in the second target data based on the weight influence factor and the initial weight.

[0014] Optionally, determining the broadband quality score of each grid cell based on the target weights includes: constructing a normalization matrix based on the second target data; constructing an optimal solution vector and a worst solution vector based on the normalization matrix, wherein the optimal solution vector consists of the maximum value of each column in the normalization matrix, and the worst solution vector consists of the minimum value of each column in the normalization matrix; calculating the sum of the distances from various index data in multiple sets of second target data to the optimal solution vector based on the target weights to obtain a first distance, and calculating the sum of the distances from various index data in multiple sets of second target data to the worst solution vector to obtain a second distance; and determining the broadband quality score of each grid cell based on the first distance and the second distance.

[0015] Optionally, after determining the broadband quality score for each grid cell, the method further includes: normalizing the broadband quality score.

[0016] According to another aspect of the embodiments of this application, a grid cell-based broadband quality dynamic evaluation device is also provided, comprising: an acquisition module, configured to acquire multiple sets of first target data for multiple grid cells within a target area, wherein each set of first target data includes: at least one type of first indicator data reflecting the quality of network equipment within the grid cell and at least one type of second indicator data reflecting the service evaluation of broadband services by users within the grid cell; a determination module, configured to determine the influencing factors affecting the first target data; a correction module, configured to correct the multiple sets of first target data according to the influencing factors to obtain multiple sets of second target data; and an evaluation module, configured to determine the target weight of each type of indicator data in the multiple sets of second target data, and determine the broadband quality score of each grid cell based on the target weight.

[0017] According to another aspect of the embodiments of this application, an electronic device is also provided, the electronic device comprising: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the above-described grid cell-based broadband quality dynamic evaluation method through the computer program.

[0018] In this embodiment, multiple sets of first target data from multiple grid cells within a target area are acquired. Each set of first target data includes at least one type of first indicator data reflecting the quality of network devices within the grid cell and at least one type of second indicator data reflecting user evaluation of broadband services within the grid cell. Influence factors affecting the first target data are determined, and the multiple sets of first target data are corrected according to these influence factors to obtain multiple sets of second target data. The target weight of each type of indicator data in the multiple sets of second target data is determined, and the broadband quality score for each grid cell is determined based on the target weight. The multiple sets of first target data constitute multi-level and multi-dimensional indicator data, comprehensively reflecting the experience of different target objects with the current broadband service. The introduction of influence factors allows for dynamic adjustment of the indicator data based on the degree of influence on different users in different scenarios, thereby making the indicator data more realistically reflect network broadband quality issues. This solves the technical problem of related technologies failing to consider multi-dimensional and multi-level indicator data, resulting in poor accuracy of broadband quality scores. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 This is a flowchart illustrating a grid cell-based dynamic evaluation method for broadband quality according to an embodiment of this application.

[0021] Figure 2 This is a tree diagram of optional broadband grid cell index data according to an embodiment of this application;

[0022] Figure 3 This is a flowchart illustrating another method for dynamic evaluation of broadband quality based on grid cells according to an embodiment of this application;

[0023] Figure 4 This is a schematic diagram of the multi-dimensional network quality scoring result of an optional mesh cell according to an embodiment of this application;

[0024] Figure 5 This is a schematic diagram of a grid cell-based broadband quality dynamic evaluation device according to an embodiment of this application. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first," "second," etc., used in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] To better understand the embodiments of this application, the following is a translation and explanation of some nouns or terms that appear in the description of the embodiments of this application:

[0028] KPI (Key Performance Indicator): These are typically important, measurable parameters that can be monitored at the network level.

[0029] KQI (Key Quality Indicator): These are business parameters that are closely related to user experience and are proposed for different business operations. They are key indicators at the business level or quality parameters for different business operations.

[0030] Grid cell: refers to the smallest unit within a grid, typically a residential area. It involves dividing the area covered by the same network into several smallest units for broadband quality evaluation.

[0031] The 3σ criterion is based on the assumption that a set of test data contains only random errors. The standard deviation is calculated and processed to obtain the standard deviation. An interval is determined according to a certain probability. Any error exceeding this interval is considered to be gross error rather than random error, and data containing such errors should be discarded.

[0032] IQR (Interquartile Range) box plot: IQR is a method for measuring variability by dividing a dataset into quartiles (i.e., dividing the data into four equal groups). Quartiles divide a rank-sorted dataset into four equal parts, where Q1 is the first quartile, Q2 is the second quartile, and Q3 is the third quartile. IQR is defined as Q3 - Q1. Data points outside the range of Q3 + 1.5 * IQR or Q3 - 1.5 * IQR are considered outliers.

[0033] Entropy method: A mathematical method used to determine the degree of dispersion of a certain indicator. The greater the dispersion of data, the greater the amount of information, the smaller the uncertainty, and the smaller the entropy value; conversely, the smaller the dispersion of data, the smaller the amount of information, the greater the uncertainty, and the larger the entropy value.

[0034] TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) algorithm: A ranking technique based on the similarity of an ideal target. It identifies the optimal and worst-performing targets among multiple evaluation objects from a normalized data matrix, calculates the distance between each evaluation object and the optimal and worst-performing targets, thus obtaining the degree of closeness between each evaluation object and the ideal target. The objects are then ranked in descending order of closeness, and this ranking is used as the basis for evaluating the relative merits of the evaluation objects.

[0035] Example 1

[0036] Because the relevant technologies do not comprehensively consider the impact of multi-dimensional and multi-level evaluation indicators on broadband quality scores, they are unable to quickly locate broadband quality problems, thus affecting user experience.

[0037] Therefore, to address the aforementioned issues, this application provides a grid-based dynamic broadband quality evaluation method. The method divides the area into several grid units for broadband quality evaluation, using residential communities as units. For each grid unit, user-based KQI data and network-based KPI data are acquired to form multi-level and multi-dimensional indicator data. This data intuitively and comprehensively reflects the experience of different target users with the current broadband service. An influencing factor is introduced to adjust the indicator data based on the degree of impact of different scenarios on different users, thereby enhancing the influence of the indicator data. This makes the indicator data more accurately reflect network broadband quality issues, thus solving the technical problem of poor accuracy in broadband quality scoring caused by related technologies that do not consider multi-dimensional and multi-level indicator data.

[0038] It should be noted that the steps shown in the flowchart in 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 may be executed in a different order than that shown here.

[0039] Figure 1 This is a flowchart illustrating an optional grid-based dynamic broadband quality evaluation method according to an embodiment of this application, as shown below. Figure 1 As shown, the method includes at least steps S102-S106, wherein:

[0040] Step S102: Obtain multiple sets of first target data from multiple grid cells within the target area. Each set of first target data includes: at least one type of first indicator data reflecting the quality of network devices within the grid cell and at least one type of second indicator data reflecting the service evaluation of broadband services by users within the grid cell.

[0041] Among them, grid units can be, but are not limited to, residential communities, and can be limited according to the actual needs of the scenario.

[0042] According to an optional embodiment of this application, network-related indicator data and user-related indicator data of multiple grid cells within the target area are first obtained; then, the network-related indicator data is preprocessed to obtain at least one type of first indicator data, wherein the first indicator data includes at least one of the following: number of failures, failure month-on-month change, failure duration, number of users affected by failure, and user quality degradation trend; finally, the user-related indicator data is preprocessed to obtain at least one type of second indicator data, wherein the second indicator data includes at least one of the following: total number of users who complain and total number of users who file complaints.

[0043] Specifically, in order to comprehensively and intuitively reflect the experience and perception of different user types regarding the current broadband service, network-related indicator data and user-related indicator data of multiple grid units within the target area can be obtained from the database. Among them, network-related indicator data mainly collects network-level data of each grid unit within the target area, while user-related indicator data mainly collects user evaluation data of different network broadband services within each grid unit.

[0044] Furthermore, to ensure the accuracy of the final broadband quality score, it is necessary to fill in missing data and remove outlier data from the network and user indicator data obtained from the database. For example, based on the business analysis of each indicator data, if there are missing network indicator data, then the corresponding indicator data should be re-obtained and the missing data filled in; alternatively, the 3σ criterion or IQR box plot can be used to detect outlier data with significant differences in the network indicator data, and these outlier data should be removed. This avoids missing and outlier data from seriously affecting the final broadband quality score.

[0045] Figure 2 A tree diagram of optional broadband grid cell indicator data is shown. After filling in missing indicators and removing outlier indicators from network and user indicator data, indicator data reflecting the broadband network quality within the grid cell can be obtained. Specifically, after preprocessing the network indicator data, primary indicator data reflecting OLT equipment faults, line faults (fault duration, fault frequency, and fault ratio) can be obtained, as well as primary indicator data reflecting poor user quality labels such as the number of broadband mismatched users and the number of users with substandard downlink optical attenuation. After preprocessing the user indicator data, secondary indicator data reflecting the total number of users who file complaints, the total number of users who receive reports, or the total number of users dissatisfied with fault conditions each month can be obtained.

[0046] After identifying multiple sets of primary target data that can reflect multiple grid cells within the target area, an influencing factor can be introduced to dynamically adjust the indicator data for different scenarios and user profiles.

[0047] Step S104: Determine the influencing factors affecting the first target data, and adjust the multiple sets of first target data according to the influencing factors to obtain multiple sets of second target data. The influencing factors include: a basic influence factor, a sensitivity factor, and a continuous deterioration factor.

[0048] Optionally, for each grid cell, the basic influence factor corresponding to the grid cell can be determined based on the first target data corresponding to the grid cell. The basic influence factor is obtained by multiplying the grid element level influence factor and the time influence factor.

[0049] Specifically, the network element level influence factor (NF) reflects the degree of influence of different types of network elements within a grid cell on users. Therefore, for each grid cell, the network element influence factor corresponding to different types of network elements within the grid cell can be determined based on a preset network element level mapping relationship. The definition of the network element level influence factor NF is as follows:

[0050] NF i =A

[0051] Here, A is a constant value. Different network element types have different network element level influence factors. The larger the value of the network element level influence factor, the higher the corresponding network element level and the wider the range of users affected. For example, A can be set to 20 for BRAS equipment, 10 for SW equipment, 3 for OLT equipment, 2 for primary optical splitter, and 1 for secondary optical splitter.

[0052] In practical applications, the specific value of A can be adjusted appropriately based on the results of the dynamic algorithm. The larger the specific value of A, the greater the impact range of network element faults.

[0053] The time impact factor (TF) reflects the degree of influence of different fault occurrence times within a grid cell on the user. Therefore, for each grid cell, the time impact factor corresponding to different fault occurrence times can be determined based on a preset fault time mapping relationship. The definition of the time impact factor (TF) is as follows:

[0054]

[0055] Wherein, B, C, D, and E are the impact coefficients for different types of users at different time periods, and all are constants. As can be seen from the above definition, different users are affected depending on when the fault occurs, and the corresponding degree of impact also varies.

[0056] In practical applications, the specific values ​​of B, C, D, and E can be adjusted appropriately based on the results of the dynamic algorithm; the larger the coefficient, the greater the influence. Additionally, the time interval value can also be adjusted.

[0057] Therefore, the basic influence factor BF can be derived from the network element level influence factor NF and the time influence factor TF. The expression for the basic influence factor BF is as follows:

[0058] BF i =NF i *TF i

[0059] Where i represents the i-th grid cell.

[0060] Optionally, sensitivity factors corresponding to the grid cells can be determined based on the first target data corresponding to the grid cells. These sensitivity factors include: complaint / report sensitivity factors, business sensitivity factors, and repetitive obstacle sensitivity factors.

[0061] Specifically, the complaint sensitivity factor reflects the impact of the number of user complaints within a grid unit on business evaluation. Therefore, for each grid unit, a first weighting coefficient for the total number of complaining users can be determined first, and then the complaint sensitivity factor can be calculated based on the total number of complaining users, the total number of users submitting complaints, and the first weighting coefficient. The specific calculation formula is as follows:

[0062]

[0063] Where, x i y represents the total number of users reporting network faults within this grid cell. i z represents the total number of users who filed complaints due to network failures within this grid cell. i This represents the total number of users in the grid cell, and α is the influence coefficient (generally greater than 1). The influence coefficient can be adjusted according to the actual situation.

[0064] The service sensitivity factor reflects the sensitivity of different broadband services within a grid cell to the network's impact and the differentiated services provided to high-value users. Therefore, for each grid cell, the number of reporting users and the number of users affected by faults for different user types can be determined first. A second impact coefficient for the number of reporting users and a third impact coefficient for the number of users affected by faults can then be determined. The service sensitivity factor is then calculated based on these factors, ensuring that at least pre-defined high-value users are included among the different user types. The specific calculation formula is as follows:

[0065]

[0066] Among them, vu i This represents the number of high-value users out of the total number of users reporting faults within that grid cell. i Gu represents the number of high-value users affected by a network failure within that grid cell. i This represents the number of game clients affected by a network failure within that grid cell. i z represents the number of live streaming clients affected by a network failure within that grid cell. i This represents the total number of users in the grid cell. x1, x2, x3, and x4 are the influence coefficients corresponding to the above variables. The influence coefficients can be adjusted according to the actual situation.

[0067] Furthermore, the repeatability sensitivity factor reflects the degree to which the number of identical network failures within a grid cell affects service evaluation. Therefore, for each grid cell, the following steps are first determined: the number of users affected by repeated failures, the total number of users affected by failures within the target evaluation period, the number of users reporting repeated failures, and the total number of users reporting repeated failures within the target evaluation period. Then, the repeatability sensitivity factor is calculated based on these factors. The specific calculation formula is as follows:

[0068]

[0069] Among them, tw i aw represents the number of users affected by the same network element failing more than twice within that grid cell. i tc represents the total number of users affected by the fault during the statistical period. i This represents the number of users within that grid cell who have reported a fault more than twice. i This represents the total number of users who submitted reports within the statistical period.

[0070] For example, network elements can usually be divided into various types such as OLT, single-outline, and double-outline. If there are many network outages of any type within a grid unit, the broadband quality score of that grid unit will be lower accordingly. Or, if the number of repeated user complaints caused by network faults in that grid unit is higher, it indicates that the network quality problem of that grid unit is more prominent.

[0071] Optionally, the continuous deterioration factor corresponding to the grid cell can also be determined based on the first target data corresponding to the grid cell. The continuous deterioration factor is used to reflect the degree of impact of the continuous state of network faults within the grid cell on the service evaluation.

[0072] Specifically, for each grid cell, the following can be determined first: the number of days of network failure duration in the first target data; the number of evaluation days in the target month; the number of first network failures in the second half of the target evaluation period; the number of second network failures in the first half of the target evaluation period; the total number of network failures; the number of third network failures and the number of first-user quality defects as of the target date; and the number of fourth network failures and the number of second-user quality defects as of the day before the target date. Then, based on the number of days of network failure duration, the number of evaluation days in the target month, the number of first network failures, the number of second network failures, the total number of network failures, the number of third network failures and the number of first-user quality defects, as well as the number of fourth network failures and the number of second-user quality defects, the continuing deterioration factor is calculated. The specific calculation formula is as follows:

[0073]

[0074] Among them, fd iThis represents the number of days the network fault lasted within that grid cell, where L represents the number of assessment days in the current month, and rw i fw represents the number of network failures in the first half of the evaluation cycle. i ft represents the number of second network failures in the first half of the evaluation period. i f represents the total number of network faults in that grid cell during the monthly assessment period. t f represents the number of third network faults up to the current grid cell. t-1 This represents the fourth network fault count for this grid cell as of yesterday, cq t This represents the percentage of users with poor performance as of now, cq t-1 This represents the percentage of users with poor quality as of yesterday.

[0075] As an optional implementation, in order to ensure that the acquired indicator data can accurately reflect the broadband quality scores of different scenarios and different types of users, the first target data can be corrected to obtain the second target data. Each set of second target data includes at least one type of third indicator data and at least one type of fourth indicator data.

[0076] Specifically, based on the influence factor, each type of first indicator data in the first target data can be corrected to obtain at least one type of third indicator data; based on the influence factor, sensitivity factor, and continuous deterioration factor, each type of second indicator data in the first target data can be corrected to obtain at least one type of fourth indicator data.

[0077] For example, regarding the duration of a failure, an impact factor is introduced to correct the indicator data, resulting in the failure impact indicator data, the expression of which is as follows:

[0078]

[0079] in, f represents the corrected fault duration category data. i Represents the actual downtime, BF i Represents the basic factor of influence, u i z represents the number of users affected by the fault. i This represents the total number of users in the community.

[0080] To correct the total number of users affected by the fault, we introduce a basic impact factor, a sensitivity factor, and a continuous deterioration factor to obtain the number of users affected by the fault, as shown in the following expression:

[0081]

[0082] in, This represents the revised metrics affecting the number of users, ui z represents the actual number of users affected by the fault. i BF represents the total number of users in this grid cell. i Representative influence factor, SF i VF represents the sensitivity of complaints and grievances. i Represents value differentiation and business sensitivity, RF i PF represents sensitivity to repetitive impairment. i This represents a factor that continues to worsen.

[0083] In addition, before determining the target weight of each type of indicator data in multiple sets of second target data, it is also necessary to perform positiveization and normalization processing on each type of indicator data in multiple sets of second target data.

[0084] Since the broadband quality evaluation system of a broadband grid community mainly includes four types of indicators: fault duration, number of users affected by a fault, number of faults, and fault-to-period ratio, the larger these indicators are for users, the worse the perceived quality. Therefore, it is necessary to use an algorithm to transform extremely small indicators into extremely large indicators to positively process the indicator data. The specific processing method is as follows:

[0085]

[0086] in, x represents the positively processed indicator data. i x represents the current value of the indicator data. max This represents the maximum value of the current indicator data.

[0087] Furthermore, in order to eliminate the influence of the dimensions of each indicator data, the original indicator data can be linearized to the range [0, 1] using a maximum-minimum linear function. The specific transformation method is as follows:

[0088]

[0089] in, The index number after normalization is ranked, x i x represents the current value of the indicator data. max x represents the maximum value of the current indicator data. min This represents the minimum value of the current indicator data.

[0090] Step S106: Determine the target weight of each type of index data in multiple sets of second target data, and determine the broadband quality score of each grid cell based on the target weight.

[0091] Typically, after determining the data for each indicator in the evaluation system, weights are set based on the experience values ​​of relevant technical personnel, and broadband network quality is calculated using a weighted method. However, this method generally lacks consideration of the impact of the continuous deterioration of individual or a small number of indicators on the final broadband quality score. As a result, it is impossible to adjust the initial weights of each indicator based on the dynamic changes in the indicator data, leading to an inability to accurately assess the overall perception of the grid unit.

[0092] To avoid the aforementioned problems, in this embodiment of the application, for each grid cell, the initial weight w of each type of index data in the second target data corresponding to the grid cell is determined using the entropy method. is The initial weights w of each category of indicator data in the second objective data are determined using a variable weighting function. is The weighting influence factors, where the weighting influence factor w id This is used to reflect the degree of change and duration of change of each type of indicator data in the second target data within the target time period, relative to the initial weight w. is The degree of influence; based on the weighted influence factor and the initial weight w is Calculate the target weight w for each type of indicator data in the second target data. ip Wherein, the target weight w ip The calculation formula is as follows:

[0093] w ip =α*w is +β*w id

[0094] In the above calculation formula, α and β represent weighting coefficients, which can be adjusted according to the actual situation, while w id This represents the weighted influence factor obtained using a variable weighting function, where the variable weighting function can be defined as:

[0095]

[0096] Where L represents the current month's evaluation period, l = 1 indicates the first evaluation period, and the value of l ranges from [1, L]. i x represents the number of users affected by the fault on that day. tl z represents the number of users who filed complaints and reports due to malfunctions on that day. i This represents the total number of users within a grid cell.

[0097] Since the target weights of the indicator data cannot be directly applied to actual calculations, it is necessary to ensure that the vector composed of the target weights of various indicator parameters satisfies... The target weights of all indicator data within the evaluation system can be normalized as follows:

[0098]

[0099] Specifically, if no network failure or user complaints or reports caused by network failure occur in the grid cell during the evaluation period, the target weight will not change; if a network failure or user complaints or reports caused by network failure occur in the grid cell during the evaluation period, the target weight will change. The longer the evaluation period, the worse the indicator degradation value of the community, the more days of continuous degradation, and the greater the target weight accordingly.

[0100] Through the above process, the target weights of various indicator data within each grid cell can be calculated. In order to combine the influence of different dimensional indicator data on broadband quality score, the final calculated broadband quality score is closer to the actual business.

[0101] In this embodiment, a normalization matrix can be constructed based on the second target data; an optimal solution vector and a worst solution vector can be constructed based on the normalization matrix, wherein the optimal solution vector consists of the maximum value of each column in the normalization matrix, and the worst solution vector consists of the minimum value of each column in the normalization matrix; based on the target weight, the sum of the distances from various index data in multiple sets of second target data to the optimal solution vector is calculated to obtain the first distance, and the sum of the distances from various index data in multiple sets of second target data to the worst solution vector is calculated to obtain the second distance; based on the first distance and the second distance, the broadband quality score of each grid cell is determined.

[0102] Specifically, the actual values ​​of various indicator data within the target area are first normalized to obtain a normalized decision matrix (i.e., a standardized matrix). The indicator data are categorized into two main types: benefit-type attributes and cost-type attributes. Therefore, the expression for the normalized decision matrix is ​​as follows:

[0103]

[0104] Among them, a ij This represents the actual value of the j-th indicator data within the i-th grid cell.

[0105] Next, based on the normalized decision matrix, the optimal solution vector c is determined. * And the worst solution vector c 0 Wherein, the optimal solution vector c * And the worst solution vector c 0 The definition is as follows:

[0106]

[0107]

[0108] Where, max i c ijmin represents the maximum value in each column of the normalized decision matrix (i.e., the maximum actual value of all indicator data within each grid cell). i c ij This represents the minimum value in each column of the normalized decision matrix (i.e., the minimum actual value of all indicator data within each grid cell).

[0109] Then, combining the target weights w = [w1, w2, ..., w] of each indicator data... n ] T The first distance is obtained by calculating the sum of the distances from various index data to the optimal solution vector in multiple sets of second target data. First distance The expression is as follows:

[0110]

[0111] Similarly, the second distance is obtained by calculating the sum of the distances from various index data to the worst solution vector in multiple sets of second target data. Second distance The expression is as follows:

[0112]

[0113] Finally, based on the first distance Second distance This allows us to obtain the broadband quality score for each grid cell. in, The expression is as follows:

[0114]

[0115] In addition, after determining the broadband quality score of each grid cell, the broadband quality score can be normalized using the maximum and minimum value method, so that the broadband quality score is mapped to [Minscore, 100]. The maximum value is set to 100, and the minimum value Minscore can be preset according to the actual evaluation results of the current network.

[0116] According to an optional embodiment of this application, the broadband quality evaluation of each grid cell can be obtained through the following steps S1-S10, wherein:

[0117] S1, retrieve network-related indicator data and user-related indicator data for multiple grid cells within the target area from the database;

[0118] S2, fill in missing indicator data and remove a field indicator data from the acquired network-type indicator data and user-type indicator data to obtain at least one type of first indicator data to reflect the quality of network equipment within the grid unit and at least one type of second indicator data to reflect the service evaluation of broadband services by users within the grid unit.

[0119] S3. Based on the first indicator data and the second indicator data, calculate the basic impact factor, the sensitivity factor, and the continuous deterioration factor respectively. The basic impact factor includes: network element level impact factor and time impact factor. The sensitivity factor includes: complaint and report sensitivity factor, business sensitivity factor, and repeated obstacle sensitivity factor.

[0120] S4. Based on the basic factors of influence, the data of the first indicator is corrected to obtain the data of the third indicator. Based on the basic factors of influence, the sensitive factors and the continuous deterioration factors, the data of the second indicator is corrected to obtain the data of the fourth indicator.

[0121] S5 performs positiveization and normalization processing on the third and fourth indicator data, respectively;

[0122] S6. The initial weight of each type of indicator data is calculated using the entropy method, and the weight influence factor of the initial weight of each type of indicator data is calculated using the variable weight function.

[0123] S7. Based on the initial weights and the corresponding weight influence factors, the target weights for each type of indicator data are obtained.

[0124] S8. Construct a normalized matrix based on the third and fourth indicator data, determine the optimal solution vector and the worst solution vector of the normalized matrix, and calculate the first distance of the sum of distances from each type of indicator data to the optimal solution vector and the second distance of the sum of distances from each type of indicator data to the worst solution vector based on the target weight of each type of indicator data.

[0125] S9. Calculate the broadband quality score for each grid cell based on the first and second distances.

[0126] Figure 4 This diagram illustrates a multi-dimensional network quality score for an optional grid cell. Through steps S1-S9, the network quality scores of 5267 grid cells in a certain community in July 2022 are analyzed across complaint reporting, equipment, line, customer quality issues, and overall dimensions. Figure 4The network quality scores for each grid unit within the community are clearly displayed across different dimensions. Verification shows that the accuracy of the obtained network quality scores is as high as 85%. Therefore, the method provided in this application embodiment facilitates relevant operators in accurately carrying out network broadband rectification on both the network and user sides, thereby achieving the goal of precisely optimizing network broadband quality.

[0127] In this embodiment, multiple sets of first target data from multiple grid cells within a target area are acquired. Each set of first target data includes at least one type of first indicator data reflecting the quality of network devices within the grid cell and at least one type of second indicator data reflecting user evaluation of broadband services within the grid cell. Influence factors affecting the first target data are determined, and the multiple sets of first target data are corrected according to these influence factors to obtain multiple sets of second target data. The target weight of each type of indicator data in the multiple sets of second target data is determined, and the broadband quality score for each grid cell is determined based on the target weight. The multiple sets of first target data constitute multi-level and multi-dimensional indicator data, comprehensively reflecting the experience of different target objects with the current broadband service. The introduction of influence factors allows for dynamic adjustment of the indicator data based on the degree of influence on different users in different scenarios, thereby making the indicator data more realistically reflect network broadband quality issues. This solves the technical problem of related technologies failing to consider multi-dimensional and multi-level indicator data, resulting in poor accuracy of broadband quality scores.

[0128] Example 2

[0129] According to an embodiment of this application, a grid-based broadband quality dynamic evaluation device for implementing the above-described grid-based broadband quality dynamic evaluation method is also provided, such as... Figure 5 As shown, the broadband quality evaluation device includes at least an acquisition module 51, a determination module 52, a correction module 53, and an evaluation module 54, wherein:

[0130] The acquisition module 51 is used to acquire multiple sets of first target data for multiple grid cells within the target area. Each set of first target data includes: at least one type of first indicator data reflecting the quality of network devices within the grid cell and at least one type of second indicator data reflecting the service evaluation of broadband services by users within the grid cell.

[0131] According to an optional embodiment of this application, the acquisition module 51 first acquires network-related indicator data and user-related indicator data of multiple grid cells within the target area; then, it preprocesses the network-related indicator data to obtain at least one type of first indicator data, wherein the first indicator data includes at least one of the following: number of failures, month-on-month failure rate, failure duration, number of users affected by failures, and user quality degradation trend; finally, it preprocesses the user-related indicator data to obtain at least one type of second indicator data, wherein the second indicator data includes at least one of the following: total number of users who complain and total number of users who file complaints.

[0132] Specifically, in order to comprehensively and intuitively reflect the experience and perception of different user types regarding the current broadband service, network-related indicator data and user-related indicator data of multiple grid units within the target area can be obtained from the database. Among them, network-related indicator data mainly collects network-level data from multiple grid units within the target area, while user-related indicator data mainly collects user evaluation data of different network broadband services within each grid unit.

[0133] Furthermore, to ensure the accuracy of the final broadband quality score, it is necessary to fill in missing data and remove outlier data from the network and user indicator data obtained from the database. For example, if missing network indicator data exists based on the business analysis of each indicator, then the corresponding indicator data should be re-obtained and the missing data filled in. Alternatively, if significant outlier data is detected in the network indicator data using the 3σ criterion or IQR box plot, then this outlier data should be removed. This prevents missing and outlier data from severely impacting the final broadband quality score.

[0134] Module 52 is used to determine the influencing factors affecting the first target data.

[0135] Optionally, for each grid cell, the determining module 52 can determine the basic influence factor corresponding to the grid cell based on the first target data corresponding to the grid cell. The basic influence factor is obtained by multiplying the grid element level influence factor and the time influence factor.

[0136] Specifically, the network element level influence factor (NF) reflects the degree of influence of different types of network elements within a grid cell on users. Therefore, for each grid cell, the network element influence factor corresponding to different types of network elements within the grid cell can be determined based on a preset network element level mapping relationship. The definition of the network element level influence factor NF is as follows:

[0137] NF i =A

[0138] Here, A is a constant value. Different network element types have different network element level influence factors. The larger the value of the network element level influence factor, the higher the corresponding network element level and the wider the range of users affected. For example, A can be set to 20 for BRAS equipment, 10 for SW equipment, 3 for OLT equipment, 2 for primary optical splitter, and 1 for secondary optical splitter.

[0139] In practical applications, the specific value of A can be adjusted appropriately based on the results of the dynamic algorithm. The larger the specific value of A, the greater the impact range of network element faults.

[0140] The time impact factor (TF) reflects the degree of influence of different fault occurrence times within a grid cell on the user. Therefore, for each grid cell, the time impact factor corresponding to different fault occurrence times can be determined based on a preset fault time mapping relationship. The definition of the time impact factor (TF) is as follows:

[0141]

[0142] Wherein, B, C, D, and E are the impact coefficients for different types of users at different time periods, and all are constants. As can be seen from the above definition, different users are affected depending on when the fault occurs, and the corresponding degree of impact also varies.

[0143] In practical applications, the specific values ​​of B, C, D, and E can be adjusted appropriately based on the results of the dynamic algorithm; the larger the coefficient, the greater the influence. Additionally, the time interval value can also be adjusted.

[0144] Therefore, the basic influence factor BF can be derived from the network element level influence factor NF and the time influence factor TF. The expression for the basic influence factor BF is as follows:

[0145] BF i =NF i *TF i

[0146] Where i represents the i-th grid cell.

[0147] Optionally, the determining module 52 can also determine the sensitivity factors corresponding to the grid cells based on the first target data corresponding to the grid cells. These sensitivity factors include: complaint / report sensitivity factors, business sensitivity factors, and repetitive obstacle sensitivity factors.

[0148] Specifically, the complaint sensitivity factor reflects the impact of the number of user complaints within a grid unit on business evaluation. Therefore, for each grid unit, a first weighting coefficient for the total number of complaining users can be determined first, and then the complaint sensitivity factor can be calculated based on the total number of complaining users, the total number of users submitting complaints, and the first weighting coefficient. The specific calculation formula is as follows:

[0149]

[0150] Where, x i y represents the total number of users reporting network faults within this grid cell. i z represents the total number of users who filed complaints due to network failures within this grid cell. i This represents the total number of users in the grid cell, and α is the influence coefficient (generally greater than 1). The influence coefficient can be adjusted according to the actual situation.

[0151] The service sensitivity factor reflects the sensitivity of different broadband services within a grid cell to the network's impact and the differentiated services provided to high-value users. Therefore, for each grid cell, the number of reporting users and the number of users affected by faults for different user types can be determined first. A second impact coefficient for the number of reporting users and a third impact coefficient for the number of users affected by faults can then be determined. The service sensitivity factor is then calculated based on these factors, ensuring that at least pre-defined high-value users are included among the different user types. The specific calculation formula is as follows:

[0152]

[0153] Among them, vu i This represents the number of high-value users out of the total number of users reporting faults within that grid cell. i Gu represents the number of high-value users affected by a network failure within that grid cell. i This represents the number of game clients affected by a network failure within that grid cell. i z represents the number of live streaming clients affected by a network failure within that grid cell. i This represents the total number of users in the grid cell. x1, x2, x3, and x4 are the influence coefficients corresponding to the above variables. The influence coefficients can be adjusted according to the actual situation.

[0154] Furthermore, the repeatability sensitivity factor reflects the degree to which the number of identical network failures within a grid cell affects service evaluation. Therefore, for each grid cell, the following steps are first determined: the number of users affected by repeated failures, the total number of users affected by failures within the target evaluation period, the number of users reporting repeated failures, and the total number of users reporting repeated failures within the target evaluation period. Then, the repeatability sensitivity factor is calculated based on these factors. The specific calculation formula is as follows:

[0155]

[0156] Among them, tw i aw represents the number of users affected by the same network element failing more than twice within that grid cell. i tc represents the total number of users affected by the fault during the statistical period. i This represents the number of users within that grid cell who have reported a fault more than twice. i This represents the total number of users who submitted reports within the statistical period.

[0157] For example, network elements can usually be divided into various types such as OLT, single-outline, and double-outline. If there are many network outages of any type within a grid unit, the broadband quality score of that grid unit will be lower accordingly. Or, if the number of repeated user complaints caused by network faults in that grid unit is higher, it indicates that the network quality problem of that grid unit is more prominent.

[0158] Optionally, the determining module 52 can also determine the continuous deterioration factor corresponding to the grid cell based on the first target data corresponding to the grid cell, wherein the continuous deterioration factor is used to reflect the degree of impact of the continuous state of network faults within the grid cell on the service evaluation.

[0159] Specifically, for each grid cell, the following can be determined first: the number of days of network failure duration in the first target data; the number of evaluation days in the target month; the number of first network failures in the second half of the target evaluation period; the number of second network failures in the first half of the target evaluation period; the total number of network failures; the number of third network failures and the number of first-user quality defects as of the target date; and the number of fourth network failures and the number of second-user quality defects as of the day before the target date. Then, based on the number of days of network failure duration, the number of evaluation days in the target month, the number of first network failures, the number of second network failures, the total number of network failures, the number of third network failures and the number of first-user quality defects, as well as the number of fourth network failures and the number of second-user quality defects, the continuing deterioration factor is calculated. The specific calculation formula is as follows:

[0160]

[0161] Among them, fd iThis represents the number of days the network fault lasted within that grid cell, where L represents the number of assessment days in the current month, and rw i fw represents the number of network failures in the first half of the evaluation cycle. i ft represents the number of second network failures in the first half of the evaluation period. i f represents the total number of network faults in that grid cell during the monthly assessment period. t f represents the number of third network faults up to the current grid cell. t-1 This represents the fourth network fault count for this grid cell as of yesterday, cq t This represents the percentage of users with poor performance as of now, cq t-1 This represents the percentage of users with poor quality as of yesterday.

[0162] The correction module 53 is used to correct multiple sets of first target data according to the influence factors to obtain multiple sets of second target data.

[0163] As an optional implementation, in order to ensure that the acquired indicator data can accurately reflect the broadband quality scores of different scenarios and different types of users, the correction module 53 can also correct each set of first target data to obtain second target data. Each set of second target data includes at least one type of third indicator data and at least one type of fourth indicator data.

[0164] Specifically, the correction module 53 can correct each type of first indicator data in the first target data based on the influence degree basic factor to obtain at least one type of third indicator data; and correct each type of second indicator data in the first target data based on the influence degree basic factor, the sensitivity factor, and the continuous deterioration factor to obtain at least one type of fourth indicator data.

[0165] For example, regarding the duration of a failure, an impact factor is introduced to correct the indicator data, resulting in the failure impact indicator data, the expression of which is as follows:

[0166]

[0167] in, f represents the corrected fault duration category data. i Represents the actual downtime, BF i Represents the basic factor of influence, u i z represents the number of users affected by the fault. i This represents the total number of users in the community.

[0168] To correct the total number of users affected by the fault, we introduce a basic impact factor, a sensitivity factor, and a continuous deterioration factor to obtain the number of users affected by the fault, as shown in the following expression:

[0169]

[0170] in, This represents the revised metrics affecting the number of users, u i z represents the actual number of users affected by the fault. i BF represents the total number of users in this grid cell. i Representative influence factor, SF i VF represents the sensitivity of complaints and grievances. i Represents value differentiation and business sensitivity, RF i PF represents sensitivity to repetitive impairment. i This represents a factor that continues to worsen.

[0171] In addition, before determining the target weight of each type of indicator data in multiple sets of second target data, it is also necessary to perform positiveization and normalization processing on each type of indicator data in multiple sets of second target data.

[0172] Since the broadband evaluation system of grid cells mainly includes four types of indicators: fault duration, number of users affected by the fault, number of faults, and fault-to-period ratio, the larger these four indicators are for users, the worse their perception becomes. Therefore, it is necessary to use an algorithm to transform extremely small indicators into extremely large indicators to positively process the indicator data. The specific processing method is as follows:

[0173]

[0174] in, x represents the positively processed indicator data. i x represents the current value of the indicator data. max This represents the maximum value of the current indicator data.

[0175] Furthermore, in order to eliminate the influence of the dimensions of each indicator data, the original indicator data can be linearized to the range [0, 1] using a maximum-minimum linear function. The specific transformation method is as follows:

[0176]

[0177] in, The index number after normalization is ranked, x i x represents the current value of the indicator data. max x represents the maximum value of the current indicator data. min Represents the minimum value of the current indicator data.

[0178] Evaluation module 54 is used to determine the target weight of each type of index data in multiple sets of second target data, and to determine the broadband quality score of each grid cell based on the target weight.

[0179] In this embodiment of the application, for each grid cell, the evaluation module 54 uses the entropy method to determine the initial weight w of each type of index data in the second target data corresponding to the grid cell. is The initial weights w of each category of indicator data in the second objective data are determined using a variable weighting function. is The weighting influence factors, where the weighting influence factor w id This is used to reflect the degree of change and duration of change of each type of indicator data in the second target data within the target time period, relative to the initial weight w. is The degree of influence; based on the weighted influence factor and the initial weight w is Calculate the target weight w for each type of indicator data in the second target data. ip Wherein, the target weight w ip The calculation formula is as follows:

[0180] w ip =α*w is +β*w id

[0181] In the above calculation formula, α and β represent weighting coefficients, which can be adjusted according to the actual situation, while w id This represents the weighted influence factor obtained using a variable weighting function, where the variable weighting function can be defined as:

[0182]

[0183] Where L represents the current month's evaluation period, l = 1 indicates the first evaluation period, and the value of l ranges from [1, L]. i x represents the number of users affected by the fault on that day. tl z represents the number of users who filed complaints and reports due to malfunctions on that day. i This represents the total number of users within a grid cell.

[0184] Since the target weights of the indicator data cannot be directly applied to actual calculations, it is necessary to ensure that the vector composed of the target weights of each indicator parameter satisfies the following conditions. The target weights of all indicator data within the evaluation system can be normalized as follows:

[0185]

[0186] Specifically, if no network failure or user complaints or reports caused by network failure occur in the grid cell during the evaluation period, the target weight will not change; if a network failure or user complaints or reports caused by network failure occur in the grid cell during the evaluation period, the target weight will change. The longer the evaluation period, the worse the indicator degradation value of the community, the more days of continuous degradation, and the greater the target weight accordingly.

[0187] Through the above process, the target weights of various indicator data within each grid cell can be calculated. In order to combine the influence of different dimensional indicator data on broadband quality score, the final calculated broadband quality score is closer to the actual business.

[0188] In this embodiment, a normalization matrix can be constructed based on the second target data; an optimal solution vector and a worst solution vector can be constructed based on the normalization matrix, wherein the optimal solution vector consists of the maximum value of each column in the normalization matrix, and the worst solution vector consists of the minimum value of each column in the normalization matrix; based on the target weight, the sum of the distances from various index data in multiple sets of second target data to the optimal solution vector is calculated to obtain the first distance, and the sum of the distances from various index data in multiple sets of second target data to the worst solution vector is calculated to obtain the second distance; based on the first distance and the second distance, the broadband quality score of each grid cell is determined.

[0189] Specifically, the actual values ​​of various indicator data within the target area are first normalized to obtain a normalized decision matrix (i.e., a standardized matrix). The indicator data are categorized into two main types: benefit-type attributes and cost-type attributes. Therefore, the expression for the normalized decision matrix is ​​as follows:

[0190]

[0191] Among them, a ij This represents the actual value of the j-th indicator data within the i-th grid cell.

[0192] Next, based on the normalized decision matrix, the optimal solution vector c is determined. * And the worst solution vector c 0 Wherein, the optimal solution vector c * And the worst solution vector c 0 The definition is as follows:

[0193]

[0194]

[0195] Where, max i c ijmin represents the maximum value in each column of the normalized decision matrix (i.e., the maximum actual value of all indicator data within each grid cell). i c ij This represents the minimum value in each column of the normalized decision matrix (i.e., the minimum actual value of all indicator data within each grid cell).

[0196] Then, combining the target weights w = [w1, w2, ..., w] of each indicator data... n ] T The first distance is obtained by calculating the sum of the distances from various index data to the optimal solution vector in multiple sets of second target data. First distance The expression is as follows:

[0197]

[0198] Similarly, the second distance is obtained by calculating the sum of the distances from various index data to the worst solution vector in multiple sets of second target data. Second distance The expression is as follows:

[0199]

[0200] Finally, based on the first distance Second distance This allows us to obtain the broadband quality score for each grid cell. in, The expression is as follows:

[0201]

[0202] In addition, after determining the broadband quality score of each grid cell, the broadband quality score can be normalized using the maximum and minimum value method, so that the broadband quality score is mapped to [Minscore, 100]. The maximum value is set to 100, and the minimum value Minscore can be preset according to the actual evaluation results of the current network.

[0203] It should be noted that each module in the grid cell-based broadband quality dynamic evaluation device in this application embodiment corresponds one-to-one with each implementation step of the grid cell-based broadband quality dynamic evaluation method in embodiment 1. Since embodiment 1 has been described in detail, some details not shown in this embodiment can be referred to embodiment 1, and will not be elaborated further here.

[0204] Example 3

[0205] According to an embodiment of this application, an electronic device is also provided, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the grid cell-based broadband quality dynamic evaluation method of Embodiment 1 through the computer program.

[0206] Specifically, the processor is configured to execute the following steps via a computer program: acquire multiple sets of first target data for multiple grid cells within a target area, wherein each set of first target data includes: at least one type of first indicator data reflecting the quality of network equipment within the grid cell and at least one type of second indicator data reflecting the service evaluation of broadband services by users within the grid cell; determine the influencing factors affecting the first target data, and correct the multiple sets of first target data according to the influencing factors to obtain multiple sets of second target data; determine the target weight of each type of indicator data in the multiple sets of second target data, and determine the broadband quality score of each grid cell based on the target weight.

[0207] Optionally, multiple sets of first target data from multiple grid cells within the target area are acquired, including: for each grid cell, acquiring network device index data and user evaluation index data within the grid cell; preprocessing the network device index data to obtain at least one type of first index data, wherein the type of the first index data includes at least one of the following: number of network device failures, month-on-month comparison of network device failures, duration of network device failures, and number of users affected by network device failures; preprocessing the user evaluation index data to obtain at least one type of second index data, wherein the type of the second index data includes at least one of the following: total number of users who complained and total number of users who filed complaints.

[0208] Optionally, the influencing factors include: a basic influence factor, a sensitive factor, and a continuous deterioration factor. Determining the influencing factors affecting the first target data includes: for each grid cell, determining the basic influence factor corresponding to the grid cell based on the first target data for that grid cell, wherein the basic influence factor is obtained by multiplying the network element influence factor (reflecting the impact of different types of network elements within the grid cell on users) and the time influence factor (reflecting the impact of different fault occurrence times within the grid cell on users); determining the sensitive factor corresponding to the grid cell based on the first target data for that grid cell, wherein the sensitive factor includes: a factor reflecting the number of user complaints within the grid cell. The complaint and complaint sensitivity factor reflects the degree of impact of different broadband services on the network within a grid unit, as well as the value difference and business sensitivity factor of differentiated services for high-value users. The repeated obstacle sensitivity factor reflects the degree of impact of the number of the same network failures within a grid unit on the business evaluation. The continuous deterioration factor is determined based on the first target data corresponding to the grid unit. The continuous deterioration factor reflects the degree of impact of the network failure persistence status within the grid unit on the business evaluation. The network failure persistence status is based on the number of network equipment failures, the duration of network equipment failures, the month-on-month comparison of network equipment failures, and the trend of user quality degradation.

[0209] Optionally, the basic factors of influence corresponding to the grid unit are determined based on the first target data corresponding to the grid unit, including: for each grid unit, determining the grid element influence factors corresponding to different types of grid elements within the grid unit based on the preset grid element level mapping relationship; and determining the time influence factors corresponding to different fault occurrence times within the grid unit based on the preset fault time mapping relationship.

[0210] Optionally, the sensitivity factors corresponding to the grid units are determined based on the first target data corresponding to the grid units, including: for each grid unit, determining a first weight coefficient for the total number of complaining users, and calculating a complaint sensitivity factor based on the total number of reporting users, the total number of complaining users, and the first weight coefficient; determining the number of reporting users corresponding to different types of users and the number of users affected by the fault corresponding to different types of users, and determining a second influence coefficient for the number of reporting users and a third influence coefficient for the number of users affected by the fault, and calculating a business sensitivity factor based on the number of reporting users, the number of users affected by the fault, the second influence coefficient, and the third influence coefficient, wherein at least pre-defined high-value users are included among the different types of users; determining the number of users affected by repeated faults, the total number of users affected by faults within the target evaluation period, the number of users who repeatedly report faults, and the total number of reporting users within the target evaluation period, and calculating a repeated obstacle sensitivity factor based on the number of users affected by repeated faults, the total number of users affected by faults, the number of users who repeatedly report faults, and the total number of reporting users.

[0211] Optionally, the continuous deterioration factor corresponding to the grid cell is determined based on the first target data corresponding to the grid cell, including: for each grid cell, determining the number of network failure duration days, the number of evaluation days in the target month, the number of first network failures in the second half of the target evaluation period, the number of second network failures in the first half of the target evaluation period, the total number of network failures, the number of third network failures and the number of first user quality defects as of the target date, the number of fourth network failures and the number of second user quality defects on the day before the target date, and calculating the continuous deterioration factor based on the number of network failure duration days, the number of evaluation days in the target month, the number of first network failures, the number of second network failures, the total number of network failures, the number of third network failures and the number of first user quality defects, as well as the number of fourth network failures and the number of second user quality defects.

[0212] Optionally, each set of second target data includes: at least one type of third indicator data and at least one type of fourth indicator data. Multiple sets of first target data are modified according to the impact factors to obtain multiple sets of second target data, including: for each set of first target data, modifying each type of first indicator data in the first target data based on the impact factor to obtain at least one type of third indicator data; modifying each type of second indicator data in the first target data based on the impact factor, the sensitivity factor, and the continuous deterioration factor to obtain at least one type of fourth indicator data.

[0213] Optionally, determining the target weight for each type of indicator data in multiple sets of second target data includes: for each grid cell, using the entropy method to determine the initial weight of each type of indicator data in the second target data corresponding to the grid cell; using a variable weight function to determine the weight influence factor of the initial weight of each type of indicator data in the second target data, wherein the weight influence factor is used to reflect the influence of the degree of change and duration of change of each type of indicator data in the second target data on the initial weight within the target time period; and calculating the target weight of each type of indicator data in the second target data based on the weight influence factor and the initial weight.

[0214] Optionally, determining the broadband quality score of each grid cell based on the target weights includes: constructing a normalization matrix based on the second target data; constructing an optimal solution vector and a worst solution vector based on the normalization matrix, wherein the optimal solution vector consists of the maximum value of each column in the normalization matrix, and the worst solution vector consists of the minimum value of each column in the normalization matrix; calculating the sum of the distances from various index data in multiple sets of second target data to the optimal solution vector based on the target weights to obtain a first distance, and calculating the sum of the distances from various index data in multiple sets of second target data to the worst solution vector to obtain a second distance; and determining the broadband quality score of each grid cell based on the first distance and the second distance.

[0215] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0216] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0217] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0218] The units described as separate components may or may not be physically separate. 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 can be selected to achieve the purpose of this embodiment according to actual needs.

[0219] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0220] If the integrated unit is implemented as 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 this application, in essence, 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. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0221] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for dynamic evaluation of broadband quality based on grid cells, characterized in that, include: Acquire multiple sets of first target data from multiple grid cells within a target area, wherein each set of first target data includes: at least one type of first indicator data reflecting the quality of network devices within the grid cell and at least one type of second indicator data reflecting the service evaluation of broadband services by users within the grid cell; The factors affecting the first target data are determined, and the first target data are corrected according to the factors to obtain the second target data. The factors include: basic influence factor, sensitive factor and continuous deterioration factor. Determine the target weight for each type of indicator data in multiple sets of the second target data, and determine the broadband quality score for each grid cell based on the target weight; The acquisition method for each group of the first target data includes: for each grid cell, acquiring network device index data and user evaluation index data within the grid cell; preprocessing the network device index data to obtain at least one type of the first index data, wherein the type of the first index data includes at least one of the following: number of network device failures, month-on-month comparison of network device failures, duration of network device failures, number of users affected by network device failures, and number of users with poor quality; and preprocessing the user evaluation index data to obtain at least one type of the second index data, wherein the type of the second index data includes at least one of the following: total number of users who complained and total number of users who filed complaints. The method for determining the influencing factors includes: for each grid cell, determining the basic influence factor corresponding to the grid cell based on the first target data corresponding to the grid cell, wherein the basic influence factor is obtained by multiplying a network element influence factor reflecting the degree of influence of different types of network elements within the grid cell on users and a time influence factor reflecting the degree of influence of different fault occurrence times within the grid cell on users; determining the sensitivity factor corresponding to the grid cell based on the first target data corresponding to the grid cell, wherein the sensitivity factor includes: a complaint sensitivity factor reflecting the degree of influence of the number of user complaints within the grid cell on service evaluation, a service sensitivity factor reflecting the degree of sensitivity of different broadband services within the grid cell to the network impact, and a repeating obstacle sensitivity factor reflecting the degree of influence of the number of identical network faults within the grid cell on broadband quality evaluation; and determining the continuous deterioration factor corresponding to the grid cell based on the first target data corresponding to the grid cell, wherein the continuous deterioration factor reflects the degree of influence of the network fault persistence status within the grid cell on service evaluation.

2. The method according to claim 1, characterized in that, Determining the basic influence factor corresponding to the grid cell based on the first target data corresponding to the grid cell includes: For each grid cell, the network element influence factor corresponding to different types of network elements within the grid cell is determined according to the preset network element level mapping relationship; The time influence factor corresponding to different fault occurrence times within the grid cell is determined based on the preset fault time mapping relationship.

3. The method according to claim 1, characterized in that, Determining the sensitivity factor corresponding to the grid cell based on the first target data corresponding to the grid cell includes: For each grid cell, a first weighting coefficient is determined for the total number of users who have filed complaints, and the complaint sensitivity factor is calculated based on the total number of users who have filed complaints, the total number of users who have filed complaints, and the first weighting coefficient. The number of reporting users and the number of users affected by the fault corresponding to different types of users are determined, and a second influence coefficient and a third influence coefficient corresponding to the number of reporting users are determined. The business sensitivity factor is calculated based on the number of reporting users, the number of users affected by the fault, the second influence coefficient, and the third influence coefficient. Among the different types of users, at least pre-defined high-value users are included. The number of users affected by repeated faults, the total number of users affected by faults within the target evaluation period, the number of users who repeatedly report faults, and the total number of users who report faults within the target evaluation period are determined. The repeated fault sensitivity factor is then calculated based on the number of users affected by repeated faults, the total number of users affected by faults, the number of users who repeatedly report faults, and the total number of users who report faults.

4. The method according to claim 1, characterized in that, Determining the continuous deterioration factor corresponding to the grid cell based on the first target data corresponding to the grid cell includes: For each grid cell, the following are determined from the first target data: the number of days of network failure duration, the number of evaluation days in the target month, the number of first network failures in the second half of the target evaluation period, the number of second network failures in the first half of the target evaluation period, the total number of network failures, the number of third network failures and the number of first user quality defects up to the target date, the number of fourth network failures and the number of second user quality defects up to the day before the target date, and the continuous deterioration factor is calculated based on the number of days of network failure duration, the number of evaluation days in the target month, the number of first network failures, the number of second network failures, the total number of network failures, the number of third network failures and the number of first user quality defects, the number of fourth network failures and the number of second user quality defects.

5. The method according to claim 1, characterized in that, Each set of second target data includes: at least one type of third indicator data and at least one type of fourth indicator data. Multiple sets of first target data are corrected according to the influencing factors to obtain multiple sets of second target data, including: For each group of the first target data, based on the influence factor, each type of the first indicator data in the first target data is corrected to obtain at least one type of the third indicator data; Based on the influence factor, the sensitivity factor, and the continuous deterioration factor, each type of the second indicator data in the first target data is modified to obtain at least one type of the fourth indicator data.

6. The method according to claim 1, characterized in that, Before determining the target weight for each type of indicator data in multiple sets of the second target data, the method further includes: For each type of indicator data in multiple sets of the second target data, positive transformation and normalization processing are performed.

7. The method according to claim 1, characterized in that, Determine the target weight for each type of indicator data in multiple sets of the second target data, including: For each grid cell, the initial weight of each type of index data in the second target data corresponding to the grid cell is determined by the entropy method. A weighting influence factor is used to determine the initial weight of each type of indicator data in the second target data using a variable weighting function. The weighting influence factor is used to reflect the influence of the degree of change and duration of change of each type of indicator data in the second target data on the initial weight within the target time period. Based on the weighting influence factor and the initial weight, calculate the target weight for each type of indicator data in the second target data.

8. The method according to claim 7, characterized in that, The broadband quality score of each grid cell is determined based on the target weight, including: Based on the second target data, construct a normalized matrix; Based on the normalization matrix, construct the optimal solution vector and the worst solution vector, wherein the optimal solution vector is composed of the maximum value of each column in the normalization matrix, and the worst solution vector is composed of the minimum value of each column in the normalization matrix; Based on the target weight, the sum of the distances from various index data in multiple sets of the second target data to the optimal solution vector is calculated to obtain the first distance, and the sum of the distances from various index data in multiple sets of the second target data to the worst solution vector is calculated to obtain the second distance; Based on the first distance and the second distance, the broadband quality score of each of the grid cells is determined.

9. The method according to claim 8, characterized in that, After determining the broadband quality score for each of the grid cells, the method further includes: The broadband quality score is then normalized.

10. A grid-based broadband quality dynamic evaluation device, characterized in that, include: The acquisition module is used to acquire multiple sets of first target data from multiple grid cells within a target area. Each set of first target data includes: at least one type of first indicator data reflecting the quality of network devices within the grid cell and at least one type of second indicator data reflecting user evaluation of broadband services within the grid cell. The acquisition method for each set of first target data includes: for each grid cell, acquiring network device indicator data and user evaluation indicator data within the grid cell; preprocessing the network device indicator data to obtain at least one type of first indicator data, wherein the type of the first indicator data includes at least one of the following: number of network device failures, month-on-month comparison of network device failures, duration of network device failures, number of users affected by network device failures, and number of users with poor quality; preprocessing the user evaluation indicator data to obtain at least one type of second indicator data, wherein the type of the second indicator data includes at least one of the following: total number of users who complain and total number of users who file complaints. A determination module is used to determine the influencing factors affecting the first target data, wherein the influencing factors include: a basic influence factor, a sensitivity factor, and a continuous deterioration factor; the determination method of the influencing factors includes: for each grid cell, determining the basic influence factor corresponding to the grid cell based on the first target data corresponding to the grid cell, wherein the basic influence factor is obtained by multiplying a network element influence factor reflecting the degree of influence of different types of network elements within the grid cell on the user and a time influence factor reflecting the degree of influence of different fault occurrence times within the grid cell on the user; determining the basic influence factor based on the first target data corresponding to the grid cell. The sensitivity factors corresponding to the grid unit include: a complaint sensitivity factor reflecting the impact of the number of user complaints within the grid unit on service evaluation; a service sensitivity factor reflecting the sensitivity of different broadband services within the grid unit to the network impact; and a repeatability sensitivity factor reflecting the impact of the number of identical network failures within the grid unit on broadband quality evaluation. The continuous deterioration factor corresponding to the grid unit is determined based on the first target data corresponding to the grid unit, wherein the continuous deterioration factor reflects the impact of the persistent network failure status within the grid unit on the service evaluation. The correction module is used to correct multiple sets of the first target data according to the influence factors to obtain multiple sets of the second target data; The evaluation module is used to determine the target weight of each type of indicator data in multiple sets of the second target data, and to determine the broadband quality score of each grid cell based on the target weight.

11. An electronic device, characterized in that, include: A memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute, via the computer program, the broadband quality dynamic evaluation method based on any one of claims 1 to 9.

Images