Data detection method and apparatus

Abstract

One embodiment of the present specification provides a data detection method and device, the method comprising: acquiring monitoring index data of a service monitoring index, reading a data distribution category of the service monitoring index according to an index type of the service monitoring index, if the data distribution category is a preset distribution category, calculating a service admission interval of the service monitoring index based on the monitoring index data, performing abnormal data detection on the monitoring index data based on the service admission interval to obtain index abnormal data of the service monitoring index, so as to provide a more accurate data detection method.

Description

TECHNICAL FIELD

[0001] The present document relates to the field of mobile communication wireless technology, and particularly relates to a data detection method and device. BACKGROUND

[0002] Mobile communication network refers to a communication medium for realizing communication between mobile users and fixed point users or between mobile users. The development of mobile communication network provides great convenience for the life and communication of users. With the more and more extensive application of mobile communication network, it is a challenge for service providers and a demand of users to provide stable and effective mobile communication services.

[0003] At present, the monitoring method based on the wireless performance index threshold is usually for cell as the monitoring object, collects the performance index data of the cell, and then filters out the cell corresponding to the performance index data greater than the index threshold in the performance index data through the index threshold. However, this method based on the wireless performance index threshold cannot be very effectively applied in some special scenarios. Therefore, there is an urgent need for a more effective method for monitoring mobile communication network. SUMMARY

[0004] An embodiment of the present specification aims to provide a data detection method and device to solve the problem of incomplete detection of index abnormal data.

[0005] To solve the above technical problems, an embodiment of the present specification is implemented as follows:

[0006] In a first aspect, an embodiment of the present specification provides a data detection method, comprising:

[0007] obtaining monitoring index data of a service monitoring index, and reading a data distribution category of the service monitoring index;

[0008] if the data distribution category is a preset distribution category, calculating a service admission interval of the service monitoring index based on the monitoring index data;

[0009] performing abnormal data detection on the monitoring index data based on the service admission interval to obtain index abnormal data of the service monitoring index.

[0010] In a second aspect, another embodiment of the present specification provides a data detection device, comprising:

[0011] a data acquisition module configured to obtain monitoring index data of a service monitoring index, and read a data distribution category of the service monitoring index;

[0012] If the data distribution category is a preset distribution category, a range calculation module is run, the range calculation module is configured to calculate a service admission range of the service monitoring index based on the monitoring index data;

[0013] A data detection module is configured to detect abnormal data of the monitoring index data based on the service admission range, to obtain index abnormal data of the service monitoring index.

[0014] In a third aspect, a further embodiment of the present specification provides a data detection device, comprising a memory, a processor, and computer executable instructions stored on the memory and executable on the processor, when the computer executable instructions are executed by the processor, the steps of the data detection method of the first aspect described above are implemented.

[0015] In a fourth aspect, a further embodiment of the present specification provides a computer readable storage medium for storing computer executable instructions, when the computer executable instructions are executed by a processor, the steps of the data detection method of the first aspect described above are implemented.

[0016] The data detection method provided by the embodiment, after obtaining the monitoring index data of the service monitoring index, first reads the data distribution category of the service monitoring index according to the index type of the service monitoring index, if the data distribution category is a preset distribution category, calculates the service admission range of the service monitoring index based on the monitoring index data, and detects the data of the monitoring index data based on the service admission range, to obtain the index abnormal data of the service monitoring index. In this way, by setting the service admission range to detect the data of the monitoring index, the accuracy and effectiveness of the obtained index abnormal data are improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the one or more embodiments of the present specification, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments described in the present specification, and those skilled in the art can also obtain other drawings according to these drawings without paying creative labor.

[0018] Figure 1 A data detection method processing flowchart is provided for an embodiment of the present specification;

[0019] Figure 2 A data detection method processing schematic diagram is provided for an embodiment of the present specification;

[0020] Figure 3A data detection method processing flowchart applied to a network index detection scene is provided for an embodiment of the present specification.

[0021] Figure 4 A data detection device schematic diagram is provided for an embodiment of the present specification.

[0022] Figure 5 A structure schematic diagram of a data detection device is provided for an embodiment of the present specification. DETAILED DESCRIPTION

[0023] In order for those skilled in the art to better understand the technical solutions in one or more embodiments of the present specification, the technical solutions in one or more embodiments of the present specification will be described clearly and completely in conjunction with the drawings in one or more embodiments of the present specification. Obviously, the described embodiments are only a part of the embodiments of the present specification, not all the embodiments. Based on one or more embodiments of the present specification, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present document.

[0024] An embodiment of a data detection method provided by the present specification is provided:

[0025] In step S102, the monitoring index data of the service monitoring index is obtained, and the data distribution category of the service monitoring index is read.

[0026] In the embodiment, the mobile communication network is monitored, and specifically, the service monitoring index refers to the performance index of the mobile communication network, such as the proportion of macro station weak coverage, the proportion of core urban low CQI (Channel Quality Indication) cell ratio, uplink user average rate, 4G (the 4th Generation mobile communication technology) network experience excellent rate and other indexes representing the performance of the mobile communication network; the monitoring index data of the service monitoring index refers to the data generated in a preset period to reflect the service monitoring index.

[0027] In this embodiment, through hypothesis testing, the calculation of the interval based on the 3σ principle can make the calculated access interval more accurate and effective to realize the detection of abnormal data of the index. Therefore, based on the Laplace criterion, the 3σ principle is used for abnormal data detection of the service monitoring index. Since the 3σ principle is used, the data detection method provided in this embodiment is used for abnormal data detection of the service monitoring index subject to normal distribution. Based on this, in this embodiment, the data distribution category of the service monitoring index is configured to determine whether the service monitoring index can use the data detection method provided in this embodiment for abnormal data detection. The data distribution category includes normal distribution and other distribution.

[0028] In specific implementation, since the monitoring index data is constantly changing, it usually has long-term and short-term fluctuation properties. The long period includes continuous growth of mobile users and continuous growth of traffic volume. The short period, for example, is the Spring Festival holiday. The traffic volume in the main urban area of a large city decreases due to the holiday return flow, while the traffic volume in the surrounding towns increases. Therefore, if the detection period is not properly set, the index abnormal data recognition will not be accurate. In order to improve the effect of data detection and avoid the influence of improper detection period setting on the accuracy of index abnormal data obtained by data detection, in an optional implementation provided in this embodiment, the process of obtaining monitoring index data of the service monitoring index is implemented in the following way:

[0029] According to the service monitoring scene where the service monitoring index is located, a detection period coefficient of the service monitoring index is determined.

[0030] Based on the detection period coefficient and the current time, the monitoring index data of the service monitoring index is read.

[0031] Specifically, different detection period coefficients are configured for different service monitoring scenes. According to the detection period coefficient of the service monitoring scene where the service monitoring index is located and the current time, the monitoring index data of the service monitoring index is read.

[0032] In the specific execution process, if fixed period sampling is used for index abnormal data detection, for example, data detection is performed once a week or once a month, the dynamic change of data cannot be reflected, and some service monitoring indexes will be affected by the change of short period traffic volume. If process period sampling data is used, the short period business characteristics will be flattened, resulting in a decrease in the sensitivity of data detection. Therefore, in this embodiment, the sampling of monitoring index data is continuous and dynamic, that is, the data of the number of days corresponding to the detection period coefficient before the current time point is used for abnormal data detection. Experiments show that a 30-day period sliding window has good effect. Optionally, the monitoring index data of the service monitoring index is obtained based on a preset period sliding window.

[0033] It should be noted that the 30-day cycle sliding window provided above is merely exemplary, and different or same detection period coefficients can be configured for different service monitoring scenarios, so as to read the monitoring index data corresponding to the detection period coefficients; if the detection period coefficient is 30 days, the monitoring index data of the past 30 days before the current time can be read.

[0034] Before reading the data distribution category of the service monitoring index, the distribution category needs to be preset in advance, and the category includes normal distribution. In an optional embodiment provided by the embodiment, the process of determining the data distribution category of the service monitoring index is implemented in the following manner:

[0035] The historical index data of the service monitoring index is input into a normal distribution verification tool for normal distribution verification, and a verification result is obtained.

[0036] To further improve the accuracy of the verification of whether the service monitoring index is subject to normal distribution, after obtaining the verification result based on the normal distribution verification tool, in an optional embodiment provided by the embodiment, in the case that the verification result is the preset distribution category and the verification result is that the service monitoring index is subject to normal distribution, secondary verification is further performed in the following manner:

[0037] In the case that the verification result is that the service monitoring index is subject to normal distribution, at least one distribution verification manner is used to verify the normal distribution of the service monitoring index.

[0038] If the verification result of each distribution verification manner in the at least one distribution verification manner is that the verification is passed, it is determined that the data distribution category of the service monitoring index is normal distribution.

[0039] Specifically, in the process of determining the data distribution category of the service monitoring index, the historical index data of the service monitoring index can be input into a normal distribution verification tool for normal distribution verification, and a verification result is obtained. In the case that the verification result is that the service monitoring index is subject to normal distribution, at least one distribution verification manner is used to verify the normal distribution of the service monitoring index. In the case that the result of the normal distribution verification of the service monitoring index by each distribution verification manner is that the service monitoring index is subject to normal distribution, it is determined that the data distribution category of the service monitoring index is normal distribution. If the normal distribution verification of the service monitoring index by any distribution verification manner fails, that is, in the case that the service monitoring index is not subject to normal distribution after verification, it is determined that the service monitoring index is not subject to normal distribution.

[0040] In the specific implementation process, when performing data distribution category inspection on the historical data of the service monitoring index, the minitab software can be used as a normal distribution inspection tool to verify the distribution form of the historical data of the service monitoring index. If the result output by the minitab software is that the service monitoring index obeys the normal distribution, then the AD (Anderson-Darling) test, the RJ (Ryan-Joiner) test, or the KS (Kolmogorov-Smirnov) test, one or more of the three distribution test methods, are used to verify whether the service monitoring index obeys the normal distribution. In the AD test, the Empirical Cumulative Distribution Function (ECDF) of the sample data is compared with the expected distribution of the normal distribution of the assumed data. The better the fitting is, the smaller the value obtained is. In the RJ test, the correlation between the data and the normal score of the data is calculated to evaluate the normality. If the correlation coefficient is close to 1, and the overall distribution is normal, the KS test is similar to the AD test, and is also an ECDF-based test. First, it is assumed that the overall data obeys the normal distribution. If the observation gap is too large, the original hypothesis that the overall data obeys the normal distribution is rejected. If any one of the above three normality test methods rejects the original hypothesis, it is considered that the data does not obey the normal distribution. If the verification results of each distribution test method in the at least one distribution test method are passed, it is determined that the data distribution category of the service monitoring index is the normal distribution.

[0041] In the specific implementation process, through historical data verification, it is known that the service monitoring index obeys the normal distribution. Subsequently, when reading the same service monitoring index data, it does not need to be verified again, and the data can be directly read and detected.

[0042] In step S104, if the data distribution category is the preset distribution category, the service admission interval of the service monitoring index is calculated based on the monitoring index data.

[0043] The preset distribution category in this embodiment refers to the normal distribution. The service admission interval refers to the value range of the monitoring index data under the condition that the data fluctuates normally, which is calculated based on the normal distribution interval (μ-3σ, μ+3σ) according to the 3σ principle.

[0044] In the specific implementation, when the monitoring index data is obtained and the monitoring index data obeys the normal distribution, the service admission interval of the service monitoring index is calculated based on the monitoring index data. In order to calculate a more accurate and effective service admission interval, in an optional implementation provided in this embodiment, the service admission interval of the service monitoring index is calculated based on the monitoring index data, which is realized in the following manner:

[0045] calculate an average value and a standard deviation of the monitoring index data, and calculate an initial upper bound and an initial lower bound of the interval based on the average value and the standard deviation;

[0046] read index constraint data of the monitoring index data, update the initial upper bound and the initial lower bound based on the index constraint data to obtain an upper bound and a lower bound of the interval;

[0047] construct the service admission interval based on the upper bound and the lower bound of the interval.

[0048] Optionally, the index constraint data of the monitoring index data is a value range of the monitoring index data pre-configured according to actual conditions.

[0049] Specifically, an average value μ and a standard deviation σ of the monitoring index data are calculated, and an initial upper bound and an initial lower bound of the interval are calculated according to (μ-3σ, μ+3σ) based on the average value and the standard deviation; index constraint data of the monitoring index data is read, and the initial upper bound and the initial lower bound are updated based on the index constraint data to obtain an upper bound and a lower bound of the interval, and the service admission interval is constructed based on the upper bound and the lower bound of the interval.

[0050] For example, if the service monitoring index is a 4G network experience excellent rate, the average value μ is 99.884 and the standard deviation σ is 0.04974 calculated by historical data, and the interval (μ-3σ, μ+3σ) of the service monitoring index is (99.7348, 100.0332), and the maximum upper limit of the 4G network experience excellent rate is 100, so the interval is updated to (99.7348, 100).

[0051] It should be noted that if the index constraint data of the monitoring index data does not exceed the upper bound of the initial interval and the lower bound of the initial interval, the upper bound and the lower bound of the interval do not need to be updated, and the service admission interval can be directly constructed according to the initial upper bound and the initial lower bound.

[0052] In step S106, based on the service admission interval, the monitoring index data is detected for abnormal data to obtain index abnormal data of the service monitoring index.

[0053] The index abnormal data of the service monitoring index refers to index data outside the monitoring index data admission interval.

[0054] In a specific implementation, if the statistical model of the Laiyida criterion is directly used for monitoring statistics, since the classical Laiyida criterion has a determination standard for index abnormal data, such as the index abnormal data in Table 1, specifically, A, B, C, D, and E are five service monitoring indexes, and 0 to 19 are the numbers of the monitoring index data, if one or more extreme values of index abnormal data appear in the sampling of an actual performance index monitoring application, it will lead to a wider range of determination of index abnormal data than when no extreme value of index abnormal data appears, that is, it will lead to inaccurate calculation of the service admission interval, such as the undetected index abnormal data in Table 2, which is affected by the extreme value of the index abnormal data, and the index abnormal data smaller than the extreme value may fall within the determination range, causing the index abnormal data smaller than the extreme value to be unable to be normally identified, and leading to a decrease in the accuracy and efficiency of the method.

[0055] A B C D E 0 8 100 2 12 0.5 1 5 125 4 14 0.4 2 45 124 3 15 3 3 7 700 2 9 0.3 4 6 102 2 10 0.2 5 2 152 3.2 11 0.3 6 3 88 1.9 13 0.3 7 75 143 3 14 5 8 4 145 4 13 0.4 9 2 170 5 12 0.3 10 5 156 4 10 0.5 11 3 169 3.5 11 0.2 12 4 89 2 13 0.2 13 6 76 2 14 0.2 14 3 800 3 12 0.8 15 5 1500 4 12 0.6 16 5 85 3 16 0.4 17 6 102 2.4 9 0.2 18 2 82 3 8 0.3 19 3 98 4 15 0.5

[0056] Table 1 index abnormal data

[0057] Affected service admission interval Indicator anomaly data Indicator anomaly data not detected A [-43.68,63.58] [A7]:75 [A2]:45 B [-811.52,1312.12] [B15]:1500 [B3]: 700, [B14]: 800 C [0.31,5.79] No No D [5.53,18.77] No No E [-2.79,4.25] [E7]:5 [E2]:3

[0058] Table 2 undetected index abnormal data

[0059] To solve the above problems, on the one hand, a monitoring and early warning method based on the fusion of a traditional examination threshold and a dynamic fluctuation statistical prediction based on the Laiyida criterion can be used; however, this method can cause the method to be unable to accurately identify due to the setting of a too large detection index. Specifically, since the data is actually in a state of continuous fluctuation in time series, the monitoring method of setting a unified detection index cannot reflect this fluctuation, and the index setting is either much better than the examination threshold or much worse than the examination threshold, and the detection effect is not accurate in actual application. On the other hand, a "cyclic evaluation elimination method" can be used to evaluate and eliminate the influence of special abnormal values. Specifically, after obtaining the data, the Laiyida criterion is first used for screening, the index abnormal data found in the first round of Laiyida criterion screening is eliminated, and then the Laiyida criterion is operated again. If no index abnormal data is found, the calculation is terminated. If index abnormal data is found again, it is eliminated and the second round of operation is performed. The above process is repeated until no index abnormal data is found.

[0060] Based on this, in an optional implementation provided by the embodiment, the service admission interval is achieved by adopting the following method for detecting abnormal data of the monitoring index data:

[0061] The first abnormal data outside the service admission interval is screened out from the monitoring index data.

[0062] If the screening result is not empty, the first abnormal data is determined as the index abnormal data of the service monitoring index.

[0063] deleting the first abnormal data in the monitoring index data, to obtain target monitoring index data;

[0064] calculating a target service access interval of the service monitoring index based on the target monitoring index data;

[0065] screening second abnormal data outside the target service access interval in the target monitoring index data; if the screening result is empty, determining that the data detection of the service monitoring index ends.

[0066]

[0067]

[0068] Table 3 Index Abnormal Data Detection Process

[0069] Final service admission interval Indicator anomaly data A [-0.97,9.75] [A7]: 75, [A2]: 45 B [21.75,214.25] [B3]: 700, [B15]: 1500, [B14]: 800 C [0.31,5.79] No D [5.53,18.77] No E [-0.13,0.86] [E7]: 5, [E2]: 3

[0070] Table 4 Index Abnormal Data Detected by Data Detection

[0071] The following is combined with reference Figure 2 The data detection method provided by the embodiment is further described. First, the monitoring index data is read. Second, the service access interval of the service monitoring data is calculated using the 3σ principle. The service monitoring data is processed using the "cyclic evaluation elimination method". Specifically, first abnormal data outside the service access interval is screened out in the monitoring index data. If the screening result is empty, it is determined that the data detection of the service monitoring index ends. If the screening result is not empty, the first abnormal data is determined as index abnormal data of the service monitoring index. The first abnormal data in the monitoring index data is deleted, to obtain target monitoring index data. The target service access interval of the service monitoring index is calculated again based on the target monitoring index data. Second abnormal data outside the target service access interval is screened out in the target monitoring index data. If the screening result is empty, it is determined that the data detection of the service monitoring index ends. If the screening result is not empty, the second abnormal data is determined as index abnormal data of the service monitoring index and the second abnormal data is deleted. The calculation of the service access interval and the screening of the index abnormal data are performed again in the above manner, until the screening result is empty. Finally, the data detection result of the service monitoring index is obtained, and the service monitoring index data detection ends.

[0072] This step uses the "cyclic evaluation elimination method" to successfully detect other index abnormal data smaller than the extreme value in the case that the monitoring index data appears an extreme value far exceeding the index abnormal data determination range, thereby improving the accuracy of index abnormal data detection.

[0073] To sum up, the data detection method provided by the embodiment first acquires monitoring index data of a service monitoring index, and reads a data distribution category of the service monitoring index. Before reading the data distribution category of the service monitoring index, a preset category needs to be set, and a normal distribution test tool is used to perform normal distribution test on historical data of the service monitoring index. In a case where the test result is that the service monitoring index is subject to normal distribution, at least one distribution test mode is used to perform normal distribution verification on the service monitoring index. When the service monitoring index is verified to be subject to normal distribution, subsequent reading of the same service monitoring index data does not need to be verified again, and the monitoring index data can be directly read for data detection.

[0074] Secondly, if the data distribution category is a preset distribution category, a mean value and a standard deviation are calculated based on the monitoring index data, an initial upper limit of an interval and an initial lower limit of the interval are calculated based on the mean value and the standard deviation, index constraint data of the monitoring index data is read, the initial upper limit of the interval and the initial lower limit of the interval are updated based on the index constraint data, an upper limit of the interval and a lower limit of the interval are obtained, and a service admission interval of the service monitoring index is constructed based on the upper limit of the interval and the lower limit of the interval. At this time, if the index constraint data does not exceed the upper limit and the lower limit of the interval, the upper limit and the lower limit of the interval do not need to be updated, and the initial upper limit and the initial lower limit of the interval are directly used as the service admission interval of the service monitoring index.

[0075] Finally, based on the service admission interval, abnormal data detection is performed on the monitoring index data, and a first abnormal data outside the service admission interval is screened out from the monitoring index data by using a “cyclic evaluation elimination method”. If the screening result is not empty, the first abnormal data is determined as index abnormal data of the service monitoring index. The first abnormal data in the monitoring index data is deleted to obtain target monitoring index data. Based on the target monitoring index data, a target service admission interval of the service monitoring index is calculated. Second abnormal data outside the target service admission interval is screened out from the target monitoring index data. If the screening result is empty, it is determined that data detection of the service monitoring index ends, and index abnormal data of the service monitoring index is obtained. In this way, by using the “cyclic evaluation elimination method”, the embodiment can detect other index abnormal data smaller than an extreme value in a case where the monitoring index data exceeds the extreme value far beyond a normal index abnormal data judgment range, can find all network abnormal fluctuation data from fluctuation performance index data, fills the theoretical gap in the field of index fluctuation analysis of performance monitoring, and improves the quality and stability of network operation. In addition, the data detection method provided by the embodiment has high implementability and strong portability, can perform fluctuation analysis data monitoring in subdivided scenes, and indirectly improves network traffic and creates economic value.

[0076] The data detection method provided in the embodiment is further described below in combination with the accompanying drawings. Figure 3 The data detection method provided in the embodiment is further described below in combination with the accompanying drawings. Figure 3 The data detection method applied to the start delay index identification scene includes steps S302 to S318.

[0077] In step S302, the monitoring index data of the monitoring index in the wireless network scene is read according to the detection period parameter of the wireless network scene.

[0078] In step S304, the data distribution category of the monitoring index is read.

[0079] In step S306, if the data distribution category of the monitoring index is normal distribution, the mean and the standard deviation of the monitoring index are calculated based on the monitoring index data.

[0080] In step S308, the service admission interval of the monitoring index is calculated according to the mean and the standard deviation.

[0081] In step S310, the index abnormal data outside the service admission interval is screened out from the monitoring index data.

[0082] Optionally, if the index abnormal data is empty, it is determined that the data detection is ended.

[0083] In step S312, if the index abnormal data is not empty, the index abnormal data is deleted from the monitoring index data to obtain target monitoring index data.

[0084] In step S314, the target service admission interval of the monitoring index is calculated based on the target monitoring index data.

[0085] In step S316, the index abnormal data of the target service admission interval is screened out from the target monitoring index data.

[0086] In step S318, if the index abnormal data is empty, it is determined that the data detection is ended.

[0087] Optionally, if the index abnormal data is not empty, steps S312 to S318 are repeated.

[0088] Figure 4 A data detection device provided in an embodiment of the application is shown in FIG. 4. Figure 4 The device includes:

[0089] The data acquisition module 402 is configured to acquire the monitoring index data of the service monitoring index, and read the data distribution category of the service monitoring index according to the index type of the service monitoring index.

[0090] The interval calculation module 404 is configured to calculate a service admission interval of the service monitoring index based on the monitoring index data if the data distribution category is a preset distribution category.

[0091] The data detection module 406 is configured to detect abnormal data of the monitoring index data based on the service admission interval, to obtain index abnormal data of the service monitoring index.

[0092] In an implementation manner, the data acquisition module 402 is specifically configured to:

[0093] According to the index type of the service monitoring index, the data is acquired by using a dynamic data acquisition method, and the data distribution category of the service monitoring index is read.

[0094] In an implementation manner, the interval calculation module 404 is specifically configured to:

[0095] If the data distribution category is a preset distribution category, the service admission interval of the service monitoring index is calculated based on the monitoring index data.

[0096] In an implementation manner, the data detection module 406 is specifically configured to:

[0097] The monitoring index data is detected for abnormal data by using a “cyclic evaluation elimination method” based on the service admission interval, to obtain index abnormal data of the service monitoring index.

[0098] The data detection method provided in the embodiment first acquires the data distribution category of the service monitoring index according to the index type of the service monitoring index by running the feature calculation module 402, and if the data distribution category is a preset distribution category, the interval calculation module 404 is run to calculate the service admission interval of the service monitoring index based on the monitoring index data. Further, after the service admission interval is obtained, the data detection module 406 is run to detect abnormal data of the monitoring index data based on the service admission interval, to obtain index abnormal data of the service monitoring index.

[0099] The data detection apparatus provided in the embodiment of the present specification can implement each process in the foregoing method embodiment, and achieve the same functions and effects, which are not repeated here.

[0100] Further, an embodiment of the present specification further provides a data detection device, Figure 5 A structural schematic diagram of the data detection device provided in an embodiment of the present specification is shown in Figure 5As shown, the device includes a memory 501, a processor 502, a bus 503 and a communication interface 504. The memory 501, the processor 502 and the communication interface 504 communicate through the bus 503, and the communication interface 504 can include an input and output interface, including but not limited to a keyboard, a mouse, a display, a microphone, a loudspeaker, etc.

[0101] Figure 5 In some embodiments, the memory 501 stores computer executable instructions that can be run on the processor 502, and the computer executable instructions are executed by the processor 502 to implement the following processes:

[0102] Obtain monitoring index data of a service monitoring index, and read a data distribution category of the service monitoring index;

[0103] If the data distribution category is a preset distribution category, calculate a service admission interval of the service monitoring index based on the monitoring index data;

[0104] Based on the service admission interval, perform abnormal data detection on the monitoring index data to obtain index abnormal data of the service monitoring index.

[0105] The data detection device provided in this embodiment, through cooperation of the memory 501, the processor 502, the bus 503 and the communication interface 504, first obtains monitoring index data of a service monitoring index, and reads a data distribution category of the service monitoring index, if the data distribution category is a preset distribution category, calculates a service admission interval of the service monitoring index based on the monitoring index data, and then based on the service admission interval, performs abnormal data detection on the monitoring index data to screen out index abnormal data outside the service interval, and re-screen target index data from which the index abnormal data is deleted, if the index abnormal data is not empty, re-delete the index abnormal data, continue to calculate a new service admission interval, and perform screening of index abnormal values, until the index abnormal data is empty, to obtain index abnormal data of the service monitoring index; in this way, data index volatility is successfully represented, and the accuracy of index abnormal data identification is improved.

[0106] The data detection device provided in an embodiment of the present specification can implement each process in the foregoing method embodiment and achieve the same functions and effects, which are not repeated here.

[0107] Further, another embodiment of the present specification further provides a computer readable storage medium for storing computer executable instructions, and the computer executable instructions are executed by a processor to implement the following processes:

[0108] acquire monitoring index data of a service monitoring index, and read a data distribution category of the service monitoring index;

[0109] if the data distribution category is a preset distribution category, calculate a service admission interval of the service monitoring index based on the monitoring index data;

[0110] based on the service admission interval, perform abnormal data detection on the monitoring index data to obtain index abnormal data of the service monitoring index.

[0111] The computer readable storage medium provided in the embodiment first acquires monitoring index data of a service monitoring index, and reads a data distribution category of the service monitoring index. If the data distribution category is a preset distribution category, a service admission interval of the service monitoring index is calculated based on the monitoring index data. Then, based on the service admission interval, abnormal data detection is performed on the monitoring index data to screen out index abnormal data outside the service interval. The target index data from which the index abnormal data is deleted is screened again. If the index abnormal data is not empty, the index abnormal data is deleted again, a new service admission interval is calculated, and screening of index abnormal values is performed, until the index abnormal data is empty, to obtain index abnormal data of the service monitoring index. In this way, data index volatility is successfully represented, and the accuracy of index abnormal data identification is improved.

[0112] The computer readable storage medium includes a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0113] The computer readable storage medium provided in the embodiment of the present specification can implement each process in the foregoing method embodiment, and achieve the same functions and effects, which are not repeated here.

[0114] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0115] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts can be implemented by computer program instructions. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts can be implemented by computer program instructions.

[0116] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts can be implemented by computer program instructions. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts can be implemented by computer program instructions.

[0117] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts can be implemented by computer program instructions. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or blocks of the flowcharts can be implemented by computer program instructions.

[0118] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0119] The memory can include non-persistent memory, random access memory (RAM), and / or non-volatile memory, such as read only memory (ROM) or flash memory, among others in a computer readable storage medium. The memory is an example of computer readable storage media.

[0120] Computer-readable storage media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable storage media does not include transient media, such as modulated data signals and carrier waves.

[0121] It should also be noted that 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. Without further limitation, 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 said element.

[0122] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0123] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A data detection method, characterized in that, include: Obtain the monitoring indicator data of the service monitoring metrics, and read the data distribution category of the service monitoring metrics; If the data distribution category is a preset distribution category, then the service access range of the service monitoring indicator is calculated based on the monitoring indicator data; Based on the service access range, abnormal data detection is performed on the monitoring indicator data to obtain abnormal data of the service monitoring indicators. The step of performing anomaly detection on the monitoring indicator data based on the service access interval to obtain the abnormal indicator data of the service monitoring indicators includes: Filter out the first abnormal data outside the service access range from the monitoring indicator data; If the filtering result is not empty, the first abnormal data is determined as the abnormal data of the service monitoring indicator; Delete the first abnormal data from the monitoring indicator data to obtain the target monitoring indicator data; Based on the target monitoring indicator data, calculate the target service access range for the service monitoring indicator; Filter out the second abnormal data outside the target service access range from the target monitoring indicator data; If the filter result is empty, then the data detection of the service monitoring indicators is considered complete.

2. The data detection method according to claim 1, characterized in that, The process of calculating the service access range for the service monitoring metrics based on the monitoring metric data includes: Calculate the average value and standard deviation of the monitoring indicator data, and calculate the upper and lower bounds of the initial interval based on the average value and standard deviation; Read the indicator constraint data of the monitoring indicator data, and update the initial upper bound and the initial lower bound of the interval based on the indicator constraint data to obtain the upper bound and the lower bound of the interval. The service access interval is constructed based on the upper and lower bounds of the interval.

3. The data detection method according to claim 1, characterized in that, The preset distribution categories include the normal distribution; The data distribution categories of the service monitoring metrics are determined in the following manner: The historical data of the service monitoring metrics are input into a normality test tool to perform a normality test and obtain the test results.

4. The data detection method according to claim 3, characterized in that, The method further includes: If the test result indicates that the service monitoring indicator follows a normal distribution, at least one distribution test method is used to verify the normality of the service monitoring indicator. If each of the at least one distribution verification method passes the verification of the service monitoring indicator, then the data distribution category of the service monitoring indicator is determined to be a normal distribution.

5. The data detection method according to claim 1, characterized in that, The monitoring metric data for obtaining service monitoring metrics includes: Based on the service monitoring scenario in which the service monitoring indicator is located, determine the detection cycle coefficient of the service monitoring indicator; Based on the detection cycle coefficient and the current time, read the monitoring indicator data of the service monitoring indicators.

6. A data detection device, characterized in that, include: The data acquisition module is configured to acquire monitoring indicator data of service monitoring indicators, and read the data distribution category of the service monitoring indicators according to the indicator type of the service monitoring indicators; If the data distribution category is a preset distribution category, then the interval calculation module is run. The interval calculation module is configured to calculate the service access interval of the service monitoring indicator based on the monitoring indicator data. The data detection module is configured to perform abnormal data detection on the monitoring indicator data based on the service access interval, and obtain the abnormal indicator data of the service monitoring indicator. The data detection module is also used to filter out the first abnormal data outside the service access range from the monitoring indicator data; If the filtering result is not empty, the first abnormal data is determined as the abnormal data of the service monitoring indicator; Delete the first abnormal data from the monitoring indicator data to obtain the target monitoring indicator data; Based on the target monitoring indicator data, calculate the target service access range for the service monitoring indicator; Filter out the second abnormal data outside the target service access range from the target monitoring indicator data; If the filter result is empty, then the data detection of the service monitoring indicators is considered complete.

7. A data detection device, characterized in that, The method includes a memory and a processor, wherein the memory stores computer-executable instructions that, when executed on the processor, enable the implementation of the steps of the method described in any one of claims 1-5.

8. A computer-readable storage medium storing computer-executable instructions, characterized in that, When the computer-executable instructions are executed by a processor, they can implement the steps of the method described in any one of claims 1-5.

Images