WO2023061342A1

WO2023061342A1 - Data trend detection method, electronic device, and storage medium

Info

Publication number: WO2023061342A1
Application number: PCT/CN2022/124465
Authority: WO
Inventors: 胡文迪
Original assignee: 中兴通讯股份有限公司
Priority date: 2021-10-13
Filing date: 2022-10-10
Publication date: 2023-04-20
Also published as: CN115982265A

Abstract

Embodiments of the present application relate to the field of communications. Disclosed are a data trend detection method, an electronic device, and a storage medium. The data trend detection method of the present application comprises: obtaining historical data and dividing same into a plurality of pieces of sub-data according to a time sequence, the historical data being data required to be responded and processed by a network device; respectively performing curve fitting on the plurality of pieces of sub-data to obtain the slope of a fitting line segment of each piece of sub-data; and performing distribution symmetry detection on the slopes of the plurality of pieces of sub-data to obtain the trend of the historical data.

Description

Data trend detection method, electronic device and storage medium

Cross References to Related Applications

This application claims priority to Chinese Patent Application No. 202111192904.5 filed on October 13, 2021, the entire content of which is hereby incorporated by reference.

technical field

The embodiment of the present application relates to the communication field, and in particular to a data trend detection method, an electronic device, and a storage medium.

Background technique

As an important infrastructure in the information age, the wireless network system needs to ensure continuous and stable operation to meet the communication needs of daily life, business and public services. To this end, it is necessary to monitor the data change trend of each component or each device in the network, and infer the current operating status of the network or a certain component based on the monitoring results.

In some cases, the Mann-Kendall trend test algorithm based on statistics is mostly used to analyze the data, but this algorithm is only effective for monotonous trend monitoring, and it is very prone to misjudgment when the data increases and decreases from time to time. , It is even impossible to judge its overall trend.

Contents of the invention

The main purpose of the embodiment of the present application is to provide a data trend detection method, electronic device and storage medium, so as to realize accurate and rapid judgment on the overall trend of data in complex situations.

In order to achieve the above purpose, an embodiment of the present application provides a data trend detection method, including: acquiring historical data, and dividing the historical data into multiple sub-data in chronological order, wherein the historical data is required to be Responding to the processed data; respectively performing curve fitting on the multiple pieces of sub-data to obtain the slope of the fitted line segment of each sub-data; performing distribution symmetry detection on the slopes of the multiple pieces of sub-data to obtain the trend of the historical data.

To achieve the above object, the embodiment of the present application also provides an electronic device, including: at least one processor; and a memory connected to the at least one processor in communication; wherein, the memory stores information that can be used by the at least one An instruction executed by a processor, the instruction is executed by the at least one processor, so that the at least one processor can execute the above data trend detection method.

To achieve the above object, the embodiments of the present application further provide a computer-readable storage medium storing a computer program, and implementing the above data trend detection method when the computer program is executed by a processor.

Description of drawings

One or more embodiments are exemplified by pictures in the accompanying drawings, and these exemplifications are not intended to limit the embodiments.

Fig. 1 is a schematic flow diagram 1 of a data trend detection method provided by an embodiment of the present application;

Fig. 2 is a schematic flow diagram II of the data trend detection method provided by the embodiment of the present application;

Fig. 3 is a schematic flow diagram three of the data trend detection method provided by the embodiment of the present application;

FIG. 4 is a flowchart of step 302 in the data trend detection method provided by the embodiment of FIG. 3 of the present application;

FIG. 5 is a schematic structural diagram of an electronic device provided in an embodiment of the present application.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the embodiments of the present application will be described in detail below with reference to the accompanying drawings. However, those of ordinary skill in the art can understand that in each embodiment of the application, many technical details are provided for readers to better understand the application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in this application can also be realized. The division of the following embodiments is for the convenience of description, and should not constitute any limitation to the specific implementation of the present application, and the embodiments can be combined and referred to each other on the premise of no contradiction.

The data trend detection method provided in this application, as shown in FIG. 1 , includes the following steps.

Step 101, acquiring historical data and dividing the historical data into multiple sub-data in chronological order, wherein the historical data is data that needs to be processed by the network device in response.

In this embodiment, the historical data is data that needs to be processed by the network device in response, such as message data. The data trend detection method of the present application can be applied to a certain service platform, can also be applied to electronic equipment in a certain network, and can also be applied to the middleware of a certain system, such as: DT (Data Technology, data technology) system middleware (such as: Kafka database).

It should be noted that the historical data mentioned in this application is a kind of real-time data. For example: on the first day, 100 pieces of data are generated, on the second day, 50 pieces of data are generated, and 40 pieces of data are processed in response, then the historical data obtained at the current moment is 110 pieces. Another example: on the first day, 100 pieces of data are generated, on the second day, no data is generated, and on the third day, no data is generated, then the historical data at the current moment is 100 pieces.

In addition, in step 101, the historical data is divided into multiple sub-data in chronological order, where the granularity of time can be one day, one hour, or one minute, one second, or even one millisecond. The specific time granularity used to divide historical data into multiple pieces of data can be determined according to the scale of the network in the application scenario and the number of network devices in the network.

In an embodiment, the historical data can be evenly divided into multiple sub-data in chronological order, and different division methods can also be selected according to the characteristics of the application scenario. For example: Since a large amount of data that needs to be responded to is generated every minute or even every second during the day, less data that needs to be responded to may be generated at night. Therefore, the historical data of the day (7:00-24:00) can be divided into multiple parts by minute, and the historical data of night (1:00-7:00) can be divided into multiple parts by hour.

In addition, before dividing the historical data into multiple pieces of sub-data in chronological order, it may further include: when missing data is detected in the historical data, filling the missing data with historical data at a moment before the missing data. That is to preprocess the historical data to ensure the integrity of the historical data.

Step 102, perform curve fitting on multiple sub-data respectively, and obtain the slope of the fitted line segment of each sub-data.

In this embodiment, when performing curve fitting on each sub-data, different fitting methods can be selected according to the distribution characteristics of each sub-data, such as least squares method, cubic spline curve fitting method, time series model Prophet, Bayeux Any existing curve fitting method such as Adams sampling method fitting to process the data, this is only an example for illustration, without any specific limitation.

In one embodiment, during the curve fitting process of two adjacent pieces of sub-data, the end point of the fitted line segment corresponding to the previous sub-data is connected to the first endpoint of the fitted line segment corresponding to the next sub-data. That is to say, when performing curve fitting on multiple sub-data to obtain the fitted line segment of each sub-data, the curve between the fitted line segments of each sub-data needs to be continuous, otherwise the slope obtained from multiple sub-data cannot accurately represent the data trend situation. For example: Divide the obtained historical data into 25 parts in chronological order, each containing 100 pieces of data, then the first sub-data is expressed as S1(k1, k2,...,k100), and the obtained fitting line segment is Y1( x), the second sub-data is expressed as S2(k1, k2,...,k100), and the obtained fitting line segment is Y2(x), then the last endpoint Y1(k100) of the first sub-data and the second sub-data The curve between the first endpoint of Y2(k1) needs to be continuous. Specifically, a point can be selected as a reference point between the end point of the first fitted line segment and the first end point of the second fitted line segment, and the fitting function of the first part of data is limited to the right, and the second The fitting function of the member data takes the limit to the left, when the right limit value of the fitting function of the first part of data at the reference point is equal to the left limit value of the fitting function of the second part of data at the reference point, it can be Ensure that there is a continuous curve between two adjacent fitting line segments. Of course, the selection method of the reference point is not limited here, it can be selected arbitrarily, and can also be selected by interpolation.

In step 103, the distribution symmetry detection is performed on the slopes of multiple sub-data to obtain the trend of historical data.

In this embodiment, when the distribution symmetry detection is performed on the slopes of multiple pieces of data, the slopes of multiple pieces of data are arranged according to the time sequence corresponding to the multiple pieces of data, so as to ensure that the real trend of historical data can be obtained. The distribution symmetry detection is performed on the slope of multiple pieces of data, that is, the symmetry center and the degree of symmetry about the symmetry center of the slope of multiple pieces of data are detected. For example: if the slope of multiple sub-data is symmetrically distributed about 0, it means that the data has no obvious trend of change. If the slope of multiple sub-data deviates from 0 and is more positive, the data shows an increasing trend. If the slope of multiple sub-data deviates from 0 and is more negative, the data shows an increasing trend.

In addition, the trend of data can pre-set a variety of trend types. For example, if the amount of data increases with time, it is an increase trend, if the amount of data decreases with time, it is a downward trend, and if the amount of data remains constant, it can be defined as The data has a tendency to accumulate (the accumulation trend can be regarded as a growth trend), or it can be defined as the data has no trend. Another example: a certain database can store up to 500 pieces of data, and currently there are 50 pieces of data, which have not been responded to and processed, and the data can be defined as no trend (neither increasing nor decreasing). Another example: a database can store up to 500 pieces of data, and at the current moment, 400 pieces of data in the database have not been responded to and processed. At this time, it can be defined that these 430 pieces of data have a tendency to accumulate or increase. That is, a threshold can be set according to the database capacity for storing data. If the amount of data is greater than the threshold, it is considered that there is an accumulation or growth trend. If the amount of data is less than the threshold, it is considered that there is no trend.

In addition, for the historical data determined to be an increasing trend, the degree of growth of the historical data may be further analyzed. For example: define a mild growth trend if the number of data growth per second is less than 1,000, and define a rapid growth trend if the number of data growth per second is greater than 5,000. Of course, this is only a specific example, and multiple different thresholds can be defined according to application scenarios and network scales to analyze and obtain the growth degree of historical data growth trends.

In one embodiment, after obtaining the historical data and before dividing the historical data into multiple sub-data in chronological order, the quantity of the historical data is obtained, and it is detected that the quantity of the historical data changes within the preset monitoring time, and the The historical data is divided into multiple sub-data in chronological order, and then

subsequent steps

102 and 103 are performed.

In one embodiment, after obtaining the historical data, it also includes: obtaining the quantity of historical data; when it is monitored that the quantity of historical data remains unchanged within the preset monitoring time and is greater than the preset fourth threshold, determine the historical data There is a growing trend. It should be noted that after the quantity of historical data is obtained, if the quantity of historical data remains unchanged, it means that the historical data has not been processed by the response. At the same time, if the amount of historical data is greater than the preset fourth threshold, it is not necessary to perform trend detection on the historical data, and directly determine that the historical data has a tendency to accumulate.

A data trend detection method proposed in this application divides historical data into multiple sub-data, and performs curve fitting on each sub-data to obtain the slope of multiple sub-data, and each slope reflects the data of a certain period of time in the historical data The trend of all slopes realizes the quantification of the overall trend of historical data. Then, the distribution symmetry detection is performed on the slope of multiple sub-data to obtain the overall trend of the historical data. The method can accurately and quickly judge the overall trend of the data in complex situations.

The embodiment of the present application relates to a data trend detection method, as shown in FIG. 2 , including the following steps.

Step 201, acquiring historical data and dividing the historical data into multiple sub-data in chronological order, wherein the historical data is data that needs to be processed by the network device in response.

It should be noted that the setting of the fourth threshold can be comprehensively determined according to the application scenario, the scale of the network, the storage capacity of the database and the needs of users. For example: if the data trend detection method of this application is applied to a network with extremely high real-time requirements , the fourth threshold can be directly set to 0, that is, if the number of historical data remains 1>0, it is determined that the data has a tendency to accumulate. Of course, this is only a specific example, and the number of the fourth thresholds may be any value.

Step 202, performing curve fitting on multiple sub-data respectively, and obtaining the slope of the fitted line segment of each sub-data.

In this embodiment, the specific implementation details of

steps

201 and 202 are basically the same as those of

steps

101 and 102, and will not be repeated here.

Step 203, check whether the symmetry center of the slope of the plurality of pieces of sub-data is 0, and obtain a first value.

In this embodiment, when testing whether the center of symmetry of the slope of the plurality of sub-data is 0, the Wilcoxon (Wilcoxon) signed rank sum test method can be used, and the Wilcoxon signed rank sum test is used to test Whether the difference between paired observations comes from a population with a mean of zero. The first value obtained is the p-value in the Wilcoxon signed rank sum test. In addition, before checking whether the center of symmetry of the slopes of the multiple pieces of sub-data is 0, the slopes of the multiple pieces of sub-data can be arranged in chronological order and combined into a slope array for processing to facilitate processing and improve calculation speed.

Step 204, performing skewness calculation on the slopes of multiple pieces of sub-data to obtain a second value.

In this embodiment, the skewness is a characteristic number that characterizes the degree of asymmetry of the probability distribution density curve relative to the average value, and it is intuitively seen as the relative length of the tail of the density function curve. For example: the skewness of the normal distribution is 0, that is, the lengths of the tails on both sides are symmetrical. When calculating the skewness of multiple sub-data, this embodiment uses the following formula to calculate:

where g ₁ is expressed as skewness, _mi is the intermediate quantity in the calculation process,

Expressed as the average value of the slope of multiple sub-data, x _n represents the slope of the n-th sub-data, and N represents the number of slope data. In this embodiment, in order to calculate the skewness of the slope of multiple molecular data with respect to 0, it can be directly replaced by 0

Calculation is performed, and the calculation result can directly represent the symmetry degree of the slope of the multi-part data with respect to 0.

Step 205, calculating the mean value and variance of the slopes of multiple pieces of sub-data, and obtaining a third value according to the mean value and variance.

In this embodiment, the result obtained by dividing the mean value of the slopes of the multiple pieces of data by the variance of the slopes of the multiple pieces of data is used as the third value.

Step 206, when the first value, the mean value of the slope array, the second value and the third value meet the preset judgment conditions, it is determined that the historical data has an increasing trend; wherein, the preset judgment conditions are the preset first conditions, At least two of the preset second condition and the preset third condition are satisfied; the first condition is that the first value is less than the preset first threshold and the average value is greater than 0, and the second condition is that the first The second value is greater than a preset second threshold, and the third condition is that the third value is greater than a preset third threshold.

In this embodiment, the first condition is represented by mathematical symbols: criteria_1=p<m1 and mean(k)>0, where p is the first value (taking the p value in the Wilcoxon signed rank sum test as an example), and m1 is The first threshold, mean(k) is the mean value of the slope of multiple sub-data; the second condition is expressed in mathematical symbols: criteria_2=g1>m2, where g1 is the skewness of multiple sub-data, and m2 is the second threshold; the third condition It is expressed in mathematical symbols as: criteria_3=[mean(k)/std(k)]>m3, where std(k) is the variance of the slope of multiple pieces of sub-data, and m3 is the third threshold. When at least two of the first condition, the second condition and the third condition are established, it is judged that the data has an increasing trend, otherwise, there is no increasing trend. It should be noted that the setting of the first threshold, the second threshold and the third threshold may be comprehensively determined according to expert experience, network scale and application scenarios. In an embodiment, the first threshold may be set to 0.05, the second threshold to 1, and the third threshold to 0.1.

In addition, when detecting the distribution symmetry of the slope of multiple pieces of data, the probability density function of the slope of multiple pieces of data can also be obtained, and the integral area (first area) of the probability density curve at x>0 and the probability density curve at x>0 can be obtained. The integral area (the second area) of the x<0 part, when the first area is greater than the second area, it can be judged that the data has a growing trend. In addition, it can also calculate the positive sum of the slopes of multiple pieces of data, calculate the negative sum of the slopes of multiple pieces of data, judge whether the ratio of the positive sum to the negative sum is greater than the threshold, and determine that the data has a growing trend when it is greater than the threshold.

In one embodiment, after determining that the historical data has a growth trend or accumulation trend, it further includes: saving the trend detection results of the historical data; re-acquiring the historical data after a preset time interval, and performing trend detection on the historical data; When the trend detection result is an increasing trend or accumulation trend again, an abnormal data accumulation alarm will be generated.

It should be noted that when the data has a growth trend or accumulation trend is determined for the first time, the detection results are saved, and after a preset time interval, the historical data is reacquired multiple times to perform trend detection on the historical data. For example: at the current moment, it is detected that the historical data has a growth trend or accumulation trend, assuming that the preset time interval is one day (24 hours), then after one day, the historical data is re-obtained, and the trend detection is performed again. When the trend detection result is an increase trend again or accumulation trend, an abnormal data accumulation alarm will be generated. Of course, when the trend detection result is an increasing trend or the number of times of the accumulation trend is multiple, an abnormal data accumulation alarm will be generated, and the specific number of times can be set arbitrarily. In addition, the accumulation trend can also be considered as a growth trend, and the specific definition of the data growth trend can be set according to the application scenario and network scale.

A data trend detection method proposed in this application divides historical data into multiple sub-data, and performs curve fitting on each sub-data to obtain the slope of multiple sub-data, and each slope reflects the data of a certain period of time in the historical data The trend of all slopes realizes the quantification of the overall trend of historical data. Then conduct Wilcoxon signed rank sum test, skewness calculation, mean and variance calculation on the slope of multiple sub-data, and accurately quantify the overall trend of historical data according to the acquired data and multiple judgment conditions. The method can accurately and quickly judge the overall trend of the data in complex situations, and avoid accurate detection when the data is piled up.

A data trend detection method proposed in this application, as shown in FIG. 3 , includes the following steps.

Step 301, acquiring the response time of historical data.

In this embodiment, the historical data is real-time data, and its response time is the moment when the amount of historical data is less than or equal to a preset fifth threshold. For example: day1, generate 100 pieces of data (historical data=100), day2, generate 50 pieces of data (historical data=150), day3, generate no data, consume 140 pieces (historical data=10<threshold, response time T1), On day4, 20 records are generated (historical data = 30), and 30 records are consumed (historical data = 0<threshold, response time T2), then two response times T1 and T2 are obtained when the historical data is processed at the current moment. It should be noted that the value of the fifth threshold may be comprehensively determined according to network scale, user requirements, application scenarios, etc., and may even be set to 0. In addition, the fourth threshold and the fifth threshold in this application may be equal or unequal.

Step 302, analyze the response time of the historical data, and obtain the analysis result of the historical data.

Further, as shown in FIG. 4 , step 302 in this embodiment may further include the following steps.

Step 401, taking the response time closest to the current moment among the response times as the first response time of the historical data.

Step 402, acquiring the time difference between the first response time and the current moment.

Step 403, comparing the time difference with a preset time threshold, and obtaining an analysis result of historical data according to the comparison result.

In this embodiment, when the time interval between the first response time and the current moment is less than a preset time threshold, it is considered that the data is responded to and processed in a timely manner; otherwise, it is considered that the data is not responded to in a timely manner. Trend detection is required.

Step 303, when the analysis result indicates that the historical data has not been responded in time, it is determined that trend detection needs to be performed on the historical data.

Step 304, acquiring historical data and dividing the historical data into multiple sub-data in chronological order, wherein the historical data is data that needs to be processed by the network device in response.

In step 305, curve fitting is performed on the plurality of pieces of sub-data respectively, and the slope of the fitted line segment of each piece of sub-data is obtained.

In step 306, the distribution symmetry detection is performed on the slopes of multiple pieces of sub-data to obtain the trend of historical data.

In this embodiment, for specific implementation details of steps 304 to 305, reference may be made to the implementation details of related steps in the above-mentioned multiple embodiments, and details are not repeated here.

A data trend detection method proposed in this application divides historical data into multiple sub-data, and performs curve fitting on each sub-data to obtain the slope of multiple sub-data, and each slope reflects the data of a certain period of time in the historical data The trend of all slopes realizes the quantification of the overall trend of historical data. Then, the distribution symmetry detection is performed on the slope of multiple sub-data to obtain the overall trend of the historical data. The method can accurately and quickly judge the overall trend of the data in complex situations. In addition, before performing trend detection on historical data, it is judged whether historical data needs to perform trend detection by analyzing the response time of historical data, so as to improve the efficiency of trend detection.

In addition, it should be understood that the division of steps in the above methods is only for clarity of description, and may be combined into one step or split into multiple steps during implementation. As long as the same logical relationship is included, all Within the scope of protection of this patent; adding insignificant modifications to the process or introducing insignificant designs without changing the core design of the process are all within the scope of protection of this patent.

The embodiment of the present application relates to an electronic device, as shown in FIG. 5 , including: at least one processor 501; and a memory 502 connected in communication with at least one processor 501; wherein, the memory 502 stores information that can be processed by at least one Instructions executed by the processor 501, the instructions are executed by at least one processor 501, so that the at least one processor 501 can execute the data trend detection method as in the above embodiment.

The electronic device includes: one or more processors 501 and a memory 502, one processor 501 is taken as an example in FIG. 5 . The processor 501 and the memory 502 may be connected through a bus or in other ways, and connection through a bus is taken as an example in FIG. 5 . The memory 502, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs and modules, such as the corresponding processing strategies in the strategy space in the embodiment of the present application The algorithm is stored in the memory 502. The processor 501 executes various functional applications and data processing of the device by running the non-volatile software programs, instructions and modules stored in the memory 502, that is, realizes the above data trend detection method.

The memory 502 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application program required by at least one function; the data storage area may store an option list and the like. In addition, the memory 502 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage devices. In some implementations, the memory 502 may optionally include a memory set remotely relative to the processor 501, and these remote memories may be connected to an external device through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

One or more modules are stored in the memory 502, and when executed by one or more processors 501, execute the data trend detection method in any method implementation above.

The above-mentioned products can execute the methods provided in the embodiments of this application, and have the corresponding functional modules and beneficial effects for executing the methods. For technical details not described in detail in this embodiment, please refer to the methods provided in the embodiments of this application.

Embodiments of the present application relate to a computer-readable storage medium storing a computer program. The above method embodiments are implemented when the computer program is executed by the processor.

That is, those skilled in the art can understand that all or part of the steps in the method of the above-mentioned embodiments can be completed by instructing related hardware through a program, the program is stored in a storage medium, and includes several instructions to make a device ( It may be a single-chip microcomputer, a chip, etc.) or a processor (processor) to execute all or part of the steps of the methods in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc., which can store program codes. .

Those of ordinary skill in the art can understand that the above-mentioned implementation modes are specific examples for realizing the present application, and in practical applications, various changes can be made to it in form and details without departing from the spirit and spirit of the present application. scope.

Claims

A data trend detection method, comprising:

Obtaining historical data, and dividing the historical data into multiple sub-data in chronological order, wherein the historical data is data that needs to be processed by the network device in response;

Curve fitting is performed on the plurality of sub-data respectively, and the slope of the fitted line segment of each sub-data is obtained;

The distribution symmetry detection is performed on the slope of multiple pieces of sub-data to obtain the trend of the historical data.
The data trend detection method according to claim 1, wherein said performing distribution symmetry detection on the slope of multiple pieces of sub-data, and obtaining the trend of said historical data comprises:

Check whether the center of symmetry of the slope of the plurality of pieces of sub-data is 0, and obtain a first value;

performing skewness calculation on the slope of the plurality of pieces of sub-data to obtain a second value;

calculating the mean and variance of the slopes of the plurality of sub-data, and obtaining a third value according to the mean and the variance;

If the first value, the average value of the slopes of the plurality of sub-data, the second value and the third value meet a preset judgment condition, it is determined that the historical data has an increasing trend;

Wherein, the preset judgment condition is that at least two conditions among the preset first condition, the preset second condition and the preset third condition are established; the first condition is that the first value is less than the preset The first threshold and the average value are greater than 0, the second condition is that the second value is greater than a preset second threshold, and the third condition is that the third value is greater than a preset third threshold.
The data trend detection method according to claim 1, wherein, after the acquisition of historical data, and before dividing the historical data into multiple sub-data in chronological order, further comprising:

The quantity of the historical data is acquired, and it is monitored that the quantity of the historical data changes within a preset monitoring time.
The data trend detection method according to claim 3, wherein, after the acquisition of historical data, further comprising:

If it is monitored that the quantity of the historical data remains unchanged within the monitoring time and is greater than the preset fourth threshold, it is determined that the historical data has an increasing trend.
The data trend detection method according to claim 1, wherein said performing curve fitting on said plurality of pieces of sub-data respectively comprises: during the process of performing curve fitting on two adjacent pieces of sub-data, the previous sub-data corresponding to The end point of the fitted line segment is connected with the first end point of the fitted line segment corresponding to the latter sub-data.
The data trend detection method according to any one of claims 1 to 5, wherein, before acquiring historical data and dividing the historical data into multiple sub-data in chronological order, further comprising:

Acquiring the response time of the historical data;

Analyzing the response time of the historical data to obtain an analysis result of the historical data;

If the analysis result indicates that the historical data has not been responded in time, it is determined that trend detection needs to be performed on the historical data.
The data trend detection method according to claim 6, wherein the response time of the historical data is the moment when the amount of historical data is less than or equal to a preset fifth threshold.
The data trend detection method according to claim 7, wherein said analyzing the response time of said historical data and obtaining the analysis result of said historical data comprises:

Taking the response time closest to the current moment among the response times as the first response time of the historical data;

Obtain the time difference between the first response time and the current moment;

The time difference is compared with a preset time threshold, and an analysis result of the historical data is obtained according to the comparison result.
The data trend detection method according to claim 2 or 4, wherein, after determining that the historical data has an increasing trend, further comprising:

saving the trend detection results of the historical data;

After a preset time interval, historical data is reacquired, and trend detection is performed on the historical data;

If the trend detection result is an increasing trend again, an abnormal data accumulation alarm is generated.
The data trend detection method according to claim 1, wherein, before dividing the historical data into multiple sub-data in chronological order, further comprising:

If it is detected that there is missing data in the historical data, the missing data is filled with historical data at a moment before the missing data.
An electronic device comprising:

at least one processor; and,

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, the instructions are executed by the at least one processor, so that the at least one processor can perform the operation described in any one of claims 1 to 10 The data trend detection method described above.
A computer-readable storage medium storing a computer program, wherein the computer program implements the data trend detection method according to any one of claims 1 to 10 when executed by a processor.