Method, device, equipment, storage medium and program product for evaluating service quality

Abstract

The application provides a service quality evaluation method, device, equipment, storage medium and program product, which are applied to the technical field of communication, and the method comprises the following steps: obtaining target service quality data of a to-be-evaluated service in a target historical time length from M terminals; performing cluster analysis on the target service quality data to determine normal values and boundary values of each service index in N service indexes in the target historical time length; determining a warning threshold value corresponding to a first service index according to the normal value and the boundary value of the first service index in the target historical time length and a warning threshold floating coefficient corresponding to the to-be-evaluated service; and evaluating real-time service quality data of the to-be-evaluated service according to the warning threshold value corresponding to each service index in the N service indexes. The method can improve the accuracy of service quality evaluation.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a service quality assessment method, apparatus, device, storage medium, and program product. Background Technology

[0002] With the application of communication technology to various industries, more stringent requirements have been placed on network service quality assurance. Poor network service quality can significantly impact production activities. Therefore, how to quickly and seamlessly assess the personalized service quality of industry clients has become a pressing issue. Currently, service quality assessment mainly relies on human experience, analyzing key service indicators such as latency and jitter, and setting fixed warning thresholds to determine whether service quality meets standards. However, different service scenarios have different network performance requirements, and fixed thresholds are insufficient to adapt to all scenarios. Furthermore, network conditions are constantly changing, and fixed warning thresholds cannot accommodate these changes, resulting in poor accuracy of existing service quality assessment methods. Summary of the Invention

[0003] This application provides a business quality assessment method, apparatus, device, storage medium, and program product to address the problem of poor accuracy in existing business quality assessment methods.

[0004] To solve the above-mentioned technical problems, this application is implemented as follows:

[0005] Firstly, embodiments of this application provide a service quality assessment method. The method includes:

[0006] Obtain target service quality data of the service to be evaluated within a target historical time period from M terminals, wherein the target service quality data includes quality data of N service indicators, and M and N are both positive integers;

[0007] Cluster analysis is performed on the target business quality data to determine the normal and boundary values ​​of each of the N business indicators within the target historical time period;

[0008] Based on the normal value and boundary value of the first business indicator within the target historical time period, and the early warning threshold fluctuation coefficient corresponding to the business to be evaluated, the early warning threshold value corresponding to the first business indicator is determined, wherein the first business indicator is any one of the N business indicators;

[0009] The real-time business quality data of the business to be evaluated is assessed based on the early warning threshold value corresponding to each of the N business indicators.

[0010] Optionally, the step of performing cluster analysis on the target business quality data to determine the normal and boundary values ​​of each of the N business indicators within the target historical time period includes:

[0011] Based on the target service quality data of the service to be evaluated for each terminal within the target historical time period, a set of service indicator feature vectors is generated to obtain M sets of service indicator feature vectors.

[0012] Density clustering is performed on the feature vectors of the M groups of business indicators to obtain L clusters, where L is a positive integer less than M;

[0013] Calculate the center point of each of the L clusters;

[0014] Based on the business indicator feature vector corresponding to the center point of each of the L clusters, determine the normal value of each of the N business indicators of the business to be evaluated within the target historical time period.

[0015] Based on the business indicator feature vector corresponding to the boundary point within each cluster, determine the boundary value of each of the N business indicators of the business to be evaluated within the target historical duration.

[0016] Optionally, the target historical duration is one month; the density clustering of the M groups of business indicator feature vectors to obtain L clusters includes:

[0017] Density clustering is performed on the feature vectors of the M groups of business indicators on a daily basis to obtain multiple clusters for each day within the target historical period;

[0018] The step of determining the normal value of each of the N business indicators of the business to be evaluated within the target historical time period based on the business indicator feature vector corresponding to the center point of each of the L clusters includes:

[0019] The weighted average of the first business indicator values ​​corresponding to the center point of each cluster in multiple clusters each day is used to determine the daily normal value of the first business indicator.

[0020] The average daily normal value of the first business indicator is determined as the monthly normal value of the first business indicator.

[0021] The step of determining the boundary value of each of the N business indicators of the business to be evaluated within the target historical duration based on the business indicator feature vector corresponding to the boundary point within each cluster includes:

[0022] The monthly boundary value of the first business indicator is determined based on the boundary point with the largest corresponding value among multiple clusters of all days included in each month.

[0023] Optionally, the weight corresponding to the center point of the j-th cluster among the multiple clusters on the i-th day is the ratio of the number of points within the j-th cluster to the number of points within all clusters on the i-th day.

[0024] Optionally, determining the warning threshold value corresponding to the first business indicator based on the normal value and boundary value of the first business indicator within the target historical time period, and the warning threshold fluctuation coefficient corresponding to the business to be evaluated, includes:

[0025] Calculate the product between the first difference and the early warning threshold fluctuation coefficient corresponding to the service to be evaluated. The first difference is the difference between the boundary value and the normal value of the first service indicator within the target historical time.

[0026] The product value is summed with the boundary value of the first business indicator within the target historical time period to determine the early warning threshold value corresponding to the first business indicator.

[0027] Optionally, the M terminals include a first terminal, which is any one of the M terminals;

[0028] The step of obtaining target service quality data for the service to be evaluated within the target historical time period from M terminals includes:

[0029] Obtain service quality data for multiple services within a target historical time period from the first terminal;

[0030] The service quality data of the multiple services within the target historical period are split to obtain the target service quality data of the service to be evaluated within the target historical period.

[0031] Secondly, embodiments of this application also provide a service quality assessment device, which includes:

[0032] The first acquisition module is used to acquire target service quality data of the service to be evaluated within a target historical time period from M terminals, wherein the target service quality data includes quality data of N service indicators, and M and N are both positive integers;

[0033] The first analysis module is used to perform cluster analysis on the target business quality data to determine the normal value and boundary value of each business indicator among the N business indicators within the target historical time period.

[0034] The first determining module is used to determine the warning threshold value corresponding to the first business indicator based on the normal value and boundary value of the first business indicator within the target historical time period, and the warning threshold fluctuation coefficient corresponding to the business to be evaluated, wherein the first business indicator is any one of the N business indicators;

[0035] The first evaluation module is used to evaluate the real-time business quality data of the business to be evaluated based on the early warning threshold value corresponding to each of the N business indicators.

[0036] Thirdly, embodiments of this application also provide an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the above-described service quality assessment method.

[0037] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described service quality assessment method.

[0038] Fifthly, a computer program product is provided, including computer instructions that, when executed by a processor, implement the steps of the method described in the first aspect.

[0039] The service quality assessment method of this application includes: acquiring target service quality data of a service to be evaluated within a target historical time period from M terminals, wherein the target service quality data includes quality data of N service indicators, where M and N are both positive integers; performing cluster analysis on the target service quality data to determine the normal value and boundary value of each of the N service indicators within the target historical time period; determining the warning threshold value corresponding to the first service indicator based on the normal value and boundary value of the first service indicator within the target historical time period and the warning threshold fluctuation coefficient corresponding to the service to be evaluated, wherein the first service indicator is any one of the N service indicators; and evaluating the real-time service quality data of the service to be evaluated based on the warning threshold values ​​corresponding to each of the N service indicators. In this method, the warning threshold value corresponding to the business indicator is determined based on the warning threshold fluctuation coefficient of the business scenario corresponding to the business indicator, as well as the normal and boundary values ​​of the business indicator in history. This method makes the determination of the warning threshold value not fixed, but dynamically adjusted with the changes in historical data and business scenarios. It can better adapt to changes in network conditions. Therefore, business quality assessment based on this warning threshold value can improve the accuracy of the assessment. Attached Figure Description

[0040] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart of the service quality assessment method provided in the embodiments of this application;

[0042] Figure 2 This is a schematic diagram of the business quality assessment system provided in the embodiments of this application;

[0043] Figure 3 This is a structural diagram of a service quality assessment device provided in an embodiment of this application;

[0044] Figure 4 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] This application provides a method for business quality assessment. See also... Figure 1 , Figure 1 This is a flowchart of the service quality assessment method provided in the embodiments of this application, such as... Figure 1 As shown, it includes the following steps:

[0047] Step 101: Obtain the target service quality data of the service to be evaluated within the target historical time period from M terminals, wherein the target service quality data includes the quality data of N service indicators, and M and N are both positive integers;

[0048] In this step, terminal probes can be used to obtain target service quality data for the service to be evaluated within a target historical timeframe from M terminals. A terminal probe is a technical tool or device used to monitor and analyze the activity of terminal devices (such as computers and mobile phones) in a network. The service to be evaluated refers to the service whose service quality needs to be assessed. The service to be evaluated may differ in different scenarios. For example, in internet companies, it might be video streaming or online gaming; in the financial industry, it might be online banking transfers or securities trading. The target historical timeframe could be the past month or the past year. The target service quality data can include quality data corresponding to various service metrics, such as latency-related quality data, jitter-related quality data, and packet loss rate-related quality data.

[0049] Step 102: Perform cluster analysis on the target business quality data to determine the normal value and boundary value of each business indicator among the N business indicators within the target historical time period;

[0050] In this step, cluster analysis is an unsupervised learning method designed to divide the dataset into different clusters, ensuring high similarity among data points within the same cluster and low similarity between data points in different clusters. Cluster analysis can quickly identify which clusters represent high-quality business operations, which represent low-quality operations, and the differences in key performance indicators (KPIs) between them. The normal values ​​of each business indicator within the target historical timeframe represent the average or typical value of that indicator, reflecting its performance during normal operation. The boundary values ​​of each business indicator within the target historical timeframe represent the extreme values ​​that the indicator may reach, reflecting special circumstances in business operation.

[0051] Step 103: Based on the normal value and boundary value of the first business indicator within the target historical time period, and the early warning threshold fluctuation coefficient corresponding to the business to be evaluated, determine the early warning threshold value corresponding to the first business indicator, wherein the first business indicator is any one of the N business indicators;

[0052] In this step, taking the first business indicator out of N business indicators as an example, a warning threshold fluctuation coefficient can be pre-determined based on the business characteristics of the business to be evaluated. Then, based on the normal and boundary values ​​of the first business indicator within the target historical time period, and the warning threshold fluctuation coefficient, the warning threshold value corresponding to the first business indicator in the current time period is determined. For example, if the business to be evaluated is an online game business and the first business indicator is latency, the warning threshold value corresponding to latency in the current month can be determined based on the normal and boundary values ​​of latency in the previous month, and the warning threshold fluctuation coefficient corresponding to the online game business.

[0053] Step 104: Evaluate the real-time business quality data of the business to be evaluated based on the early warning threshold value corresponding to each of the N business indicators.

[0054] In this step, based on the warning threshold value corresponding to each business indicator, the real-time business quality data corresponding to each business indicator of the business to be evaluated is assessed. If the real-time business quality data exceeds this warning threshold value, it means that there may be a problem with the business to be evaluated. For example, the warning threshold value corresponding to the latency indicator of the online game business is 40ms, while the obtained real-time business quality data corresponding to the latency indicator of the online game business is 60ms, which indicates that there is a performance problem in the online game business, which may cause players to experience latency, stuttering, or other negative impacts during the game.

[0055] In one implementation, target service quality data of the service to be evaluated within a target historical time period is obtained from M terminals. The target service quality data includes quality data for N service indicators, where M and N are both positive integers. Cluster analysis is performed on the target service quality data to determine the normal and boundary values ​​of each of the N service indicators within the target historical time period. Based on the normal and boundary values ​​of a first service indicator within the target historical time period, and the early warning threshold fluctuation coefficient corresponding to the service to be evaluated, an early warning threshold value corresponding to the first service indicator is determined, where the first service indicator is any one of the N service indicators. The real-time service quality data of the service to be evaluated is then assessed based on the early warning threshold values ​​corresponding to each of the N service indicators. In this implementation, the warning threshold value corresponding to the business indicator is determined based on the warning threshold fluctuation coefficient of the business scenario corresponding to the business indicator, as well as the normal and boundary values ​​of the business indicator in history. This method makes the determination of the warning threshold value not fixed, but dynamically adjusted with the changes in historical data and business scenarios. It can better adapt to changes in network conditions. Therefore, conducting business quality assessment based on this warning threshold value can improve the accuracy of the assessment.

[0056] Optionally, the step of performing cluster analysis on the target business quality data to determine the normal and boundary values ​​of each of the N business indicators within the target historical time period includes:

[0057] Based on the target service quality data of the service to be evaluated for each terminal within the target historical time period, a set of service indicator feature vectors is generated to obtain M sets of service indicator feature vectors.

[0058] Density clustering is performed on the feature vectors of the M groups of business indicators to obtain L clusters, where L is a positive integer less than M;

[0059] Calculate the center point of each of the L clusters;

[0060] Based on the business indicator feature vector corresponding to the center point of each of the L clusters, determine the normal value of each of the N business indicators of the business to be evaluated within the target historical time period.

[0061] Based on the business indicator feature vector corresponding to the boundary point within each cluster, determine the boundary value of each of the N business indicators of the business to be evaluated within the target historical duration.

[0062] In one implementation, the DBSCAN algorithm is used to perform density clustering on M groups of business indicator feature vectors. The specific process of density clustering is as follows: the similarity between points is calculated based on Euclidean distance, and then the points are divided into multiple clusters based on the similarity between points, so that the similarity within a cluster is as high as possible, while the similarity between clusters is as low as possible. Therefore, the algorithm will group similar feature vectors into the same cluster. Of course, not all business indicator feature vectors will be classified into independent clusters; some business indicator vectors may not be clustered together because they are relatively abnormal.

[0063] The feature vector of the business indicator corresponding to the center point of each cluster can be roughly understood as the average value of the feature vectors of all business indicators within that cluster. The normal value, in essence, represents the general performance of the business indicator. Therefore, based on the feature vector of the business indicator corresponding to the center point of each of the L clusters, the normal value of each of the N business indicators of the business to be evaluated within the target historical time period can be determined. However, since the DBSCAN algorithm cannot directly obtain the center point of the cluster, it needs to be combined with the K-means algorithm to calculate the center point of each cluster as the basis for determining whether the business indicator is in a normal state. Furthermore, it should be noted that the feature vector of the business indicator is a feature vector formed by the target business quality data corresponding to the N business indicators, while the final normal value is the normal value corresponding to a single business indicator.

[0064] The feature vector of the business metric corresponding to the boundary point within each cluster can generally be understood as the most "extreme" data point within the cluster (such as the point furthest from the center point), while the boundary value is essentially used to characterize the maximum or minimum value that the business metric may appear within the normal business range. Furthermore, it should be noted that the feature vector of the business metric is a feature vector formed by the target business quality data corresponding to N business metrics, while the final boundary value is the boundary value corresponding to a single business metric.

[0065] In this implementation, the feature vectors of business metrics obtained through density clustering analysis can effectively identify the service quality performance of different terminals within a specific historical period. First, the generated feature vectors provide a basis for the actual performance of each terminal, making the evaluation more targeted. Density clustering groups similar data points into the same cluster, helping to discover potential business patterns or abnormal behaviors, thereby improving the accuracy of anomaly detection. Calculating the center points of the clusters helps determine the normal values ​​of each business metric, making the evaluation criteria more consistent with reality. Furthermore, by identifying the boundary points of each cluster, reasonable boundary values ​​for business metrics can be defined, thus better reflecting the normal range of service performance.

[0066] Optionally, the target historical duration is one month; the density clustering of the M groups of business indicator feature vectors to obtain L clusters includes:

[0067] Density clustering is performed on the feature vectors of the M groups of business indicators on a daily basis to obtain multiple clusters for each day within the target historical period;

[0068] The step of determining the normal value of each of the N business indicators of the business to be evaluated within the target historical time period based on the business indicator feature vector corresponding to the center point of each of the L clusters includes:

[0069] The weighted average of the first business indicator values ​​corresponding to the center point of each cluster in multiple clusters each day is used to determine the daily normal value of the first business indicator.

[0070] The average daily normal value of the first business indicator is determined as the monthly normal value of the first business indicator.

[0071] The step of determining the boundary value of each of the N business indicators of the business to be evaluated within the target historical duration based on the business indicator feature vector corresponding to the boundary point within each cluster includes:

[0072] The monthly boundary value of the first business indicator is determined based on the boundary point with the largest corresponding value among multiple clusters of all days included in each month.

[0073] In one implementation, each set of business indicator feature vectors is based on hourly units. Therefore, if clusters are formed on a daily basis, density clustering of the business indicator feature vectors corresponding to each terminal across multiple terminals over 24 hours is required to obtain multiple clusters for each day. By using feature vectors on an hourly basis, the changing trends of business indicators within a day can be fully captured. This is because hourly data can more accurately reflect subtle differences in user behavior, avoiding misleading results caused by daily average data masking short-term fluctuations.

[0074] The daily normal value of the first business indicator is determined by taking a weighted average of the values ​​of the first business indicator corresponding to the center point of each cluster in multiple clusters each day. See the following calculation formula (1) for details:

[0075]

[0076] Among them, V D Indicates the daily normal value. This represents the first business indicator value corresponding to the center point within the i-th cluster. K represents the weight percentage of the index for the i-th cluster, and K represents the total number of clusters per day.

[0077] The average daily normal value of the first business indicator is determined as the monthly normal value of the first business indicator. For details, please refer to the following calculation formula (2):

[0078]

[0079] Among them, V M This represents the normal monthly value. H represents the daily normal value on day j, and H represents the number of days in a month.

[0080] The monthly boundary value of the first business indicator is determined based on the boundary point with the largest value among multiple clusters covering all days in a month. For example, if multiple clusters are formed for each day, and a month includes 30 days, the boundary point with the largest value among these multiple clusters can be determined as the monthly boundary value of the first business indicator. See formula (3) below for details:

[0081] V E =max(V)(3)

[0082] Among them, V E V represents the monthly boundary value, and V represents the first business indicator value within all clusters included each month.

[0083] In this implementation, the density clustering method based on daily data can effectively identify and analyze normal fluctuations and abnormal situations in business indicators, which is beneficial for the subsequent accurate evaluation of real-time business quality data.

[0084] Optionally, the weight corresponding to the center point of the j-th cluster among the multiple clusters on the i-th day is the ratio of the number of points within the j-th cluster to the number of points within all clusters on the i-th day.

[0085] In this implementation, by calculating weights, the importance of different clusters within the overall data can be clearly defined. Considering weights when evaluating business performance avoids relying solely on a small number of data points, enhancing the stability and accuracy of the evaluation. This allows the evaluation results to more accurately reflect the overall business situation.

[0086] Optionally, determining the warning threshold value corresponding to the first business indicator based on the normal value and boundary value of the first business indicator within the target historical time period, and the warning threshold fluctuation coefficient corresponding to the business to be evaluated, includes:

[0087] Calculate the product between the first difference and the early warning threshold fluctuation coefficient corresponding to the service to be evaluated. The first difference is the difference between the boundary value and the normal value of the first service indicator within the target historical time.

[0088] The product value is summed with the boundary value of the first business indicator within the target historical time period to determine the early warning threshold value corresponding to the first business indicator.

[0089] In one implementation, the warning threshold value corresponding to the first business indicator is determined based on the normal value and boundary value of the first business indicator within the target historical time period, and the warning threshold fluctuation coefficient corresponding to the business to be evaluated. For details, please refer to the following calculation formula (4):

[0090] θ=α·(V E -V M )+V E (4)

[0091] Where θ represents the warning threshold value corresponding to the first business indicator, α represents the warning threshold fluctuation coefficient corresponding to the business to be evaluated, and V E V represents the boundary value of the first business indicator within the target historical time period. M This represents the normal value of the first business indicator within the target historical time period.

[0092] In this implementation, the warning threshold value corresponding to the business indicator is determined based on the warning threshold fluctuation coefficient of the business scenario corresponding to the business indicator, as well as the normal and boundary values ​​of the business indicator in history. This method makes the determination of the warning threshold value not fixed, but dynamically adjusted with the changes in historical data and business scenarios. It can better adapt to changes in network conditions, thereby improving the accuracy of business quality assessment based on this warning threshold value.

[0093] For example, latency and jitter data of a terminal probe of an Automated Guided Vehicle (AGV) were collected last month. The indicators were then clustered using the DBSCAN and K-means algorithms to obtain a scatter plot of clustered data for a given day, as shown below. Figure 2 As shown, the scatter plot of the clustering represents latency on the horizontal axis and jitter on the vertical axis. The calculated daily normal latency value is 24ms, and the daily boundary latency value is 46ms. By statistically analyzing the daily normal and boundary values ​​for all days within a month, the monthly normal latency value is calculated to be 25ms, and the monthly boundary latency value is 47ms. The warning threshold fluctuation coefficient corresponding to the AGV is 1.5, therefore the calculated corresponding warning threshold value is:

[0094] 1.5 × (47ms - 25ms) + 47ms = 80ms

[0095] Optionally, the M terminals include a first terminal, which is any one of the M terminals;

[0096] The step of obtaining target service quality data for the service to be evaluated within the target historical time period from M terminals includes:

[0097] Obtain service quality data for multiple services within a target historical time period from the first terminal;

[0098] The service quality data of the multiple services within the target historical period are split to obtain the target service quality data of the service to be evaluated within the target historical period.

[0099] In one implementation, service quality data for multiple services within a target historical period is obtained from a first terminal. For example, if the service of the first terminal in the current month needs to be evaluated, service quality data from the previous month can be collected. The service quality data for multiple services within the target historical period are then categorized according to service type to obtain initial service quality data for the service to be evaluated within the target historical market. This initial service quality data is then cleaned by removing duplicate records and missing values ​​to obtain the target service quality data.

[0100] In this implementation, classifying the collected business data ensures that the analysis focuses on specific business processes, avoiding reduced processing efficiency due to excessive data volume. Furthermore, data cleaning, by removing duplicate records and missing values, significantly improves data accuracy, ensuring that subsequent analysis results are not affected by noise, thereby enhancing the credibility of decisions.

[0101] See Figure 3 , Figure 3 This is a structural diagram of a service quality assessment device provided in one embodiment of this application. Figure 3 As shown, the business quality assessment device 300 includes:

[0102] The first acquisition module 301 is used to acquire target service quality data of the service to be evaluated within a target historical time period from M terminals, wherein the target service quality data includes quality data of N service indicators, and M and N are both positive integers;

[0103] The first analysis module 302 is used to perform cluster analysis on the target business quality data to determine the normal value and boundary value of each business indicator among the N business indicators within the target historical time period.

[0104] The first determining module 303 is used to determine the warning threshold value corresponding to the first business indicator based on the normal value and boundary value of the first business indicator within the target historical time period, and the warning threshold floating coefficient corresponding to the business to be evaluated, wherein the first business indicator is any one of the N business indicators;

[0105] The first evaluation module 304 is used to evaluate the real-time business quality data of the business to be evaluated based on the early warning threshold value corresponding to each of the N business indicators.

[0106] Optionally, the first analysis module includes:

[0107] The first generation unit is used to generate a set of service indicator feature vectors based on the target service quality data of the service to be evaluated for each terminal within the target historical time period, thereby obtaining M sets of service indicator feature vectors.

[0108] The first clustering unit is used to perform density clustering on the M groups of business indicator feature vectors to obtain L clusters, where L is a positive integer less than M;

[0109] The first calculation unit is used to calculate the center point of each of the L clusters;

[0110] The first determining unit is used to determine the normal value of each of the N business indicators of the business to be evaluated within the target historical time period based on the business indicator feature vector corresponding to the center point of each of the L clusters.

[0111] The second determining unit is used to determine the boundary value of each of the N business indicators of the business to be evaluated within the target historical duration based on the business indicator feature vector corresponding to the boundary point in each cluster.

[0112] Optionally, the target historical duration is one month; the first clustering unit includes:

[0113] The first clustering subunit is used to perform density clustering on the feature vectors of the M groups of business indicators on a daily basis to obtain multiple clusters for each day within the target historical period.

[0114] The first determining unit includes:

[0115] The first determining subunit is used to perform a weighted average of the first business indicator values ​​corresponding to the center point of each of the multiple clusters each day, and to determine the daily normal value of the first business indicator.

[0116] The second determining subunit is used to determine the average value of the daily normal value of the first business indicator as the monthly normal value of the first business indicator.

[0117] The second determining unit includes:

[0118] The third determining subunit is used to determine the monthly boundary value of the first business indicator based on the boundary point with the largest corresponding first business indicator value among multiple clusters of all days included in the month.

[0119] Optionally, the weight corresponding to the center point of the j-th cluster among the multiple clusters on the i-th day is the ratio of the number of points within the j-th cluster to the number of points within all clusters on the i-th day.

[0120] Optionally, the first determining module includes:

[0121] The first calculation unit is used to calculate the product value between the first difference and the early warning threshold floating coefficient corresponding to the business to be evaluated. The first difference is the difference between the boundary value and the normal value of the first business indicator within the target historical time.

[0122] The third determining unit is used to sum the product value and the boundary value of the first business indicator within the target historical time period to determine the early warning threshold value corresponding to the first business indicator.

[0123] Optionally, the M terminals include a first terminal, which is any one of the M terminals;

[0124] The first acquisition module includes:

[0125] The first acquisition unit is used to acquire service quality data of multiple services within a target historical time period from the first terminal;

[0126] The first splitting unit is used to split the business quality data of the multiple services within the target historical time period, and obtain the target business quality data of the service to be evaluated among the multiple services within the target historical time period.

[0127] This application also provides an electronic device. Since the principle by which the electronic device solves the problem is similar to the service quality assessment method in this application, the implementation of this electronic device can be found in the implementation of the method, and repeated details will not be described again. Figure 4 As shown, the electronic device according to an embodiment of this application includes: a processor 400, configured to read a program from a memory 420 and execute the following processes:

[0128] Obtain target service quality data of the service to be evaluated within a target historical time period from M terminals, wherein the target service quality data includes quality data of N service indicators, and M and N are both positive integers;

[0129] Cluster analysis is performed on the target business quality data to determine the normal and boundary values ​​of each of the N business indicators within the target historical time period.

[0130] Based on the normal value and boundary value of the first business indicator within the target historical time period, and the early warning threshold fluctuation coefficient corresponding to the business to be evaluated, the early warning threshold value corresponding to the first business indicator is determined, wherein the first business indicator is any one of the N business indicators;

[0131] The real-time business quality data of the business to be evaluated is assessed based on the early warning threshold value corresponding to each of the N business indicators.

[0132] Optionally, the processor 400 is configured to read the program from the memory 420 and execute the following processes:

[0133] Based on the target service quality data of the service to be evaluated for each terminal within the target historical time period, a set of service indicator feature vectors is generated to obtain M sets of service indicator feature vectors.

[0134] Density clustering is performed on the feature vectors of the M groups of business indicators to obtain L clusters, where L is a positive integer less than M;

[0135] Calculate the center point of each of the L clusters;

[0136] Based on the business indicator feature vector corresponding to the center point of each of the L clusters, determine the normal value of each of the N business indicators of the business to be evaluated within the target historical time period.

[0137] Based on the business indicator feature vector corresponding to the boundary point within each cluster, determine the boundary value of each of the N business indicators of the business to be evaluated within the target historical duration.

[0138] Optionally, the period of the target historical duration is the target historical duration itself, which is one month; the processor 400 is configured to read the program from the memory 420 and execute the following processes:

[0139] Density clustering is performed on the feature vectors of the M groups of business indicators on a daily basis to obtain multiple clusters for each day within the target historical period;

[0140] The processor 400 is used to read the program in the memory 420 and execute the following processes:

[0141] The weighted average of the first business indicator values ​​corresponding to the center point of each cluster in multiple clusters each day is used to determine the daily normal value of the first business indicator.

[0142] The average daily normal value of the first business indicator is determined as the monthly normal value of the first business indicator.

[0143] The processor 400 is used to read the program in the memory 420 and execute the following processes:

[0144] The monthly boundary value of the first business indicator is determined based on the boundary point with the largest corresponding value among multiple clusters of all days included in each month.

[0145] Optionally, the weight corresponding to the center point of the j-th cluster among the multiple clusters on the i-th day is the ratio of the number of points within the j-th cluster to the number of points within all clusters on the i-th day.

[0146] Optionally, the processor 400 is configured to read the program from the memory 420 and execute the following processes:

[0147] Calculate the product between the first difference and the early warning threshold fluctuation coefficient corresponding to the service to be evaluated. The first difference is the difference between the boundary value and the normal value of the first service indicator within the target historical time.

[0148] The product value is summed with the boundary value of the first business indicator within the target historical time period to determine the early warning threshold value corresponding to the first business indicator.

[0149] Optionally, the M terminals include a first terminal, which is any one of the M terminals;

[0150] The processor 400 is used to read the program in the memory 420 and execute the following processes:

[0151] Obtain service quality data for multiple services within a target historical time period from the first terminal;

[0152] The service quality data of the multiple services within the target historical period are split to obtain the target service quality data of the service to be evaluated within the target historical period.

[0153] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described service quality assessment method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0154] This application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the above-described... Figure 1 The various processes of the method embodiments shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0155] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0156] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0157] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A business quality assessment method, characterized in that, The method includes: Obtain target service quality data of the service to be evaluated within a target historical time period from M terminals, wherein the target service quality data includes quality data of N service indicators, and M and N are both positive integers; Cluster analysis is performed on the target business quality data to determine the normal and boundary values ​​of each of the N business indicators within the target historical time period. Based on the normal value and boundary value of the first business indicator within the target historical time period, and the early warning threshold fluctuation coefficient corresponding to the business to be evaluated, the early warning threshold value corresponding to the first business indicator is determined, wherein the first business indicator is any one of the N business indicators; The real-time business quality data of the business to be evaluated is assessed based on the early warning threshold value corresponding to each of the N business indicators.

2. The business quality assessment method according to claim 1, characterized in that, The cluster analysis of the target business quality data to determine the normal and boundary values ​​of each of the N business indicators within the target historical time period includes: Based on the target service quality data of the service to be evaluated for each terminal within the target historical time period, a set of service indicator feature vectors is generated to obtain M sets of service indicator feature vectors. Density clustering is performed on the feature vectors of the M groups of business indicators to obtain L clusters, where L is a positive integer less than M; Calculate the center point of each of the L clusters; Based on the business indicator feature vector corresponding to the center point of each of the L clusters, determine the normal value of each of the N business indicators of the business to be evaluated within the target historical time period. Based on the business indicator feature vector corresponding to the boundary point within each cluster, determine the boundary value of each of the N business indicators of the business to be evaluated within the target historical duration.

3. The business quality assessment method according to claim 2, characterized in that, The target historical duration is one month; the density clustering of the M groups of business indicator feature vectors yields L clusters, including: Density clustering is performed on the feature vectors of the M groups of business indicators on a daily basis to obtain multiple clusters for each day within the target historical period; The step of determining the normal value of each of the N business indicators of the business to be evaluated within the target historical time period based on the business indicator feature vector corresponding to the center point of each of the L clusters includes: The weighted average of the first business indicator values ​​corresponding to the center point of each cluster in multiple clusters each day is used to determine the daily normal value of the first business indicator. The average daily normal value of the first business indicator is determined as the monthly normal value of the first business indicator. The step of determining the boundary value of each of the N business indicators of the business to be evaluated within the target historical duration based on the business indicator feature vector corresponding to the boundary point within each cluster includes: The monthly boundary value of the first business indicator is determined based on the boundary point with the largest corresponding value among multiple clusters of all days included in each month.

4. The business quality assessment method according to claim 3, characterized in that, The weight corresponding to the center point of the j-th cluster among multiple clusters on day i is the ratio of the number of points in the j-th cluster to the number of points in all clusters on day i.

5. The business quality assessment method according to claim 1, characterized in that, The step of determining the warning threshold value corresponding to the first business indicator based on the normal value and boundary value of the first business indicator within the target historical time period, and the warning threshold fluctuation coefficient corresponding to the business to be evaluated, includes: Calculate the product between the first difference and the early warning threshold fluctuation coefficient corresponding to the service to be evaluated. The first difference is the difference between the boundary value and the normal value of the first service indicator within the target historical time. The product value is summed with the boundary value of the first business indicator within the target historical time period to determine the early warning threshold value corresponding to the first business indicator.

6. The business quality assessment method according to claim 1, characterized in that, The M terminals include a first terminal, which is any one of the M terminals; The step of obtaining target service quality data for the service to be evaluated within the target historical time period from M terminals includes: Obtain service quality data for multiple services within a target historical time period from the first terminal; The service quality data of the multiple services within the target historical period are split to obtain the target service quality data of the service to be evaluated within the target historical period.

7. A business quality assessment device, characterized in that, The device includes: The first acquisition module is used to acquire target service quality data of the service to be evaluated within a target historical time period from M terminals, wherein the target service quality data includes quality data of N service indicators, and M and N are both positive integers; The first analysis module is used to perform cluster analysis on the target business quality data to determine the normal value and boundary value of each business indicator among the N business indicators within the target historical time period. The first determining module is used to determine the warning threshold value corresponding to the first business indicator based on the normal value and boundary value of the first business indicator within the target historical time period, and the warning threshold fluctuation coefficient corresponding to the business to be evaluated, wherein the first business indicator is any one of the N business indicators; The first evaluation module is used to evaluate the real-time business quality data of the business to be evaluated based on the early warning threshold value corresponding to each of the N business indicators.

8. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the business quality assessment method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the business quality assessment method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes computer instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 6.

Images