CN115801635A - An information detection method, device, equipment and readable storage medium - Google Patents

Abstract

The application discloses an information detection method, an information detection device, information detection equipment and a readable storage medium, and relates to the technical field of communication, so that the accuracy of network micro-burst detection is improved. The method comprises the following steps: acquiring first network delay data; extracting feature data of the first network delay data by using a feature extraction algorithm; processing the characteristic data by using a data detection algorithm to obtain a processing result; and determining whether the network micro burst occurs according to the processing result. The method and the device can improve the accuracy of the detection of the network micro-burst.

Description

An information detection method, device, equipment and readable storage medium

technical field

The present application relates to the technical field of communication, and in particular to an information detection method, device, equipment and readable storage medium.

Background technique

A network microburst (Microburst) means that a network element port receives a very large amount of burst data within a very short period of time (millisecond level). The duration of a typical network microburst is usually between 1 and 100 milliseconds, so that the instantaneous burst rate reaches dozens or hundreds of times the average rate, and even exceeds the port bandwidth.

There are three main ways to detect network microbursts: real-time detection of device traffic rate, real-time detection of network delay, and real-time detection of network packet loss.

Among them, the method of detecting network delay in real time has a higher accuracy rate. When a network micro-burst occurs, the device cache increases, and the processing delay will also increase. However, the occurrence time and duration of the network micro-burst are irregular, and the network delay is related to the network topology, the number of device hops, and the device type. Especially when the network scale is large, the delay value changes greatly. Therefore, how to detect the delay mutation value and duration caused by network microbursts in the real-time changing delay data is more critical. Among them, a common solution is to realize the detection of delay mutation by setting a fixed threshold, that is, when the delay mutation value exceeds the threshold, the delay mutation can be regarded as delay mutation data, and then the network microburst is determined.

However, this method may cause misjudgment of network microbursts. For example, because the time delay during normal operation of the network is also variable, and the range of change is not fixed, especially when the network path is switched, the network time delay will change greatly. However, the delay change of network path switching lasts for a long time and will not recover immediately, so it is not a delay mutation caused by network microbursts. If a fixed threshold is set, the delay change caused by path switching may be misjudged as a network microburst.

Therefore, there is a need for a method for improving the detection accuracy of network microbursts.

Contents of the invention

Embodiments of the present application provide an information detection method, device, device, and readable storage medium, so as to improve the detection accuracy of network microbursts.

In the first aspect, the embodiment of the present application provides an information detection method, including:

Obtain the first network delay data;

Using a feature extraction algorithm to extract feature data of the first network delay data;

Using a data detection algorithm to process the feature data to obtain a processing result;

According to the processing result, it is determined whether a network microburst occurs.

Wherein, the feature extraction algorithm includes LSTM (Long Short-Term Memory, long short-term memory);

The extraction of feature data of the first network delay data using a feature extraction algorithm includes:

Using LSTM to preprocess the first network delay data to obtain second network delay data;

Using the LSTM to perform feature extraction on the second network delay data to obtain the feature data.

Wherein, the feature extraction of the second network delay data by using the LSTM to obtain the feature data includes:

Using the first layer neural network of the LSTM to perform pooling processing on the second network delay data to obtain intermediate data;

The second-layer neural network of the LSTM is used to perform pooling processing on the intermediate data to obtain the feature data.

Wherein, the data detection algorithm includes: VAE (Variational Auto Encoder variational self-encoding); the use of the data detection algorithm to process the feature data includes:

Using VAE to detect the feature data to obtain third network delay data;

Comparing the feature data with the third network delay data to obtain a comparison result;

The determining whether a network microburst occurs according to the processing result includes:

When the comparison result indicates that the feature data is different from the third network delay data, it is determined that a network microburst occurs.

Wherein, the data detection algorithm includes: VAE; after the feature data is processed by using the data detection algorithm and the processing result is obtained, the method further includes:

Using an intermediate variable of VAE to obtain a first score value, wherein the intermediate variable is generated during the process of detecting the feature data using VAE;

Utilize SVR (Support Vector Regression, support vector regression) to detect described intermediate variable, obtain the second score value;

The determining whether a network microburst occurs according to the processing result includes:

When the first score value is greater than the second score value, it is determined that a network microburst occurs.

Wherein, the method also includes:

Use historical score values and historical intermediate variables to train the SVR model;

Wherein, the historical score value is obtained by using VAE to detect the characteristic data of the historical network delay data, and the historical intermediate variable is generated during the process of using the VAE to detect the characteristic data of the historical network delay data.

In the second aspect, the embodiment of the present application also provides an information detection device, including:

The first obtaining module is used to obtain the first network delay data;

The first extraction module, the module uses a feature extraction algorithm to extract the feature data of the first network delay data;

The first processing module is used to process the feature data by using a data detection algorithm to obtain a processing result;

The first determining module is configured to determine whether a network microburst occurs according to the processing result.

Wherein, the feature extraction algorithm includes LSTM; the first extraction module includes:

A preprocessing submodule, configured to use LSTM to preprocess the first network delay data to obtain second network delay data;

The feature extraction submodule is configured to use the LSTM to perform feature extraction on the second network delay data to obtain the feature data.

Wherein, the feature extraction submodule includes:

The first processing unit is configured to use the first layer neural network of the LSTM to perform pooling processing on the delay data of the second network to obtain intermediate data;

The second processing unit is configured to use the second layer neural network of the LSTM to perform pooling processing on the intermediate data to obtain the feature data.

Wherein, the data detection algorithm includes: VAE; the first processing module is configured to compare the feature data with the third network delay data to obtain a comparison result;

The first determining module is configured to determine that a network microburst occurs when the comparison result indicates that the feature data is different from the third network delay data.

Wherein, the data detection algorithm includes: VAE; the device also includes:

The second acquisition module is used to obtain the first score value by using the intermediate variable of VAE, wherein the intermediate variable is generated during the process of detecting the characteristic data by using VAE

The third acquisition module is used to detect the intermediate variable by using support vector regression SVR to obtain the second score value;

The first determining module is configured to determine that a network microburst occurs when the first score value is greater than the second score value.

Wherein, the device also includes:

The training module is used to train the SVR model by utilizing historical score values and historical intermediate variables;

Wherein, the historical score value is obtained by using VAE to detect the characteristic data of the historical network delay data, and the historical intermediate variable is generated during the process of using the VAE to detect the characteristic data of the historical network delay data.

In a third aspect, the embodiment of the present application also provides an information detection device, including: a processor and a transceiver;

Wherein, the processor is used for:

Obtain the first network delay data;

Using a feature extraction algorithm to extract feature data of the first network delay data;

Using a data detection algorithm to process the feature data to obtain a processing result;

According to the processing result, it is determined whether a network microburst occurs.

Wherein, the feature extraction algorithm includes LSTM; the processor is used for:

Using LSTM to preprocess the first network delay data to obtain second network delay data;

Using the LSTM to perform feature extraction on the second network delay data to obtain the feature data.

Wherein, the processor is used for:

Using the first layer neural network of the LSTM to perform pooling processing on the second network delay data to obtain intermediate data;

The second-layer neural network of the LSTM is used to perform pooling processing on the intermediate data to obtain the feature data.

Wherein, the data detection algorithm includes: VAE; the processor is used for:

Using VAE to detect the feature data to obtain third network delay data;

Comparing the feature data with the third network delay data to obtain a comparison result;

The determining whether a network microburst occurs according to the processing result includes:

When the comparison result indicates that the feature data is different from the third network delay data, it is determined that a network microburst occurs.

Wherein the data detection algorithm includes: VAE; the processor is used for:

Using an intermediate variable of VAE to obtain a first score value, wherein the intermediate variable is generated during the process of detecting the feature data using VAE;

Using SVR to detect the intermediate variable to obtain a second score value;

The determining whether a network microburst occurs according to the processing result includes:

When the first score value is greater than the second score value, it is determined that a network microburst occurs.

Wherein, the processor is used for:

Use historical score values and historical intermediate variables to train the SVR model;

Wherein, the historical score value is obtained by using VAE to detect the characteristic data of the historical network delay data, and the historical intermediate variable is generated during the process of using the VAE to detect the characteristic data of the historical network delay data.

In the fourth aspect, the embodiment of the present application also provides an electronic device, including: a memory, a processor, and a program stored in the memory and operable on the processor. When the processor executes the program, the above-mentioned Steps in the information detection method.

In a fifth aspect, the embodiment of the present application further provides a readable storage medium, where a program is stored on the readable storage medium, and when the program is executed by a processor, the steps in the information detection method as described above are implemented.

In the embodiment of the present application, feature extraction algorithm is used to extract the feature data of the acquired first network delay data, and the feature data is detected by data detection algorithm to obtain the processing result. It can be seen from this that in the embodiment of the present application, the feature data is processed through the data detection algorithm, and it is not necessary to set a fixed threshold to judge the network microburst as in the prior art, thereby avoiding network microbursts. Misjudgment of micro-bursts can improve the detection accuracy of network micro-bursts.

Description of drawings

Fig. 1 is one of the flowcharts of the information detection method provided by the embodiment of the present application;

Fig. 2 is the second flow chart of the information detection method provided by the embodiment of the present application;

Fig. 3 is the third flowchart of the information detection method provided by the embodiment of the present application;

Fig. 4 is one of the structural diagrams of the information detection device provided by the embodiment of the present application;

FIG. 5 is the second structural diagram of the information detection device provided by the embodiment of the present application.

Detailed ways

The term "and/or" in the embodiments of this application describes the association relationship of associated objects, indicating that there may be three relationships, for example, A and/or B, which may mean: A exists alone, A and B exist simultaneously, and B exists alone These three situations. The character "/" generally indicates that the contextual objects are an "or" relationship.

The term "plurality" in the embodiments of the present application refers to two or more, and other quantifiers are similar.

The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

Referring to Fig. 1, Fig. 1 is a flow chart of the information detection method provided by the embodiment of the present application, as shown in Fig. 1, including the following steps:

Step 101. Acquire first network delay data.

Wherein, the first network delay data refers to original network delay data, including jitter, packet loss, bandwidth and so on. Here, the first network delay data may be obtained by means of the prior art, for example, the first network delay data provided by the network delay monitoring device and the like may be obtained.

Step 102. Extract feature data of the first network delay data by using a feature extraction algorithm.

In the embodiment of the present application, the feature extraction algorithm includes LSTM. As one of the important algorithms in the field of deep learning, LSTM uses the "gate" structure to remove or increase the flow capacity of information, and can use one or more memory units with internal states to understand the contextual relationship in real-time input time-delay data. to study.

The first network delay data collected in real time is stored in a time series database. Afterwards, in this step, the first network delay data can be preprocessed using LSTM to obtain second network delay data, and the LSTM can be used to perform feature extraction on the second network delay data to obtain the obtained feature data. Wherein, the preprocessing includes processing missing data, data redundancy and data imbalance, and the like.

In the embodiment of the present application, a two-layer neural network is used in the LSTM to extract feature data, thereby making the extracted feature data more accurate. Specifically, using the first layer neural network of the LSTM to perform pooling processing on the second network delay data to obtain intermediate data, and using the second layer neural network of the LSTM to perform pooling processing on the intermediate data , to obtain the characteristic data. That is to say, the output of the first-layer neural network will still be recalculated by the second-layer neural network, so that more accurate feature data can be obtained and the accuracy of network microburst detection can be improved. Wherein, the structures of the first layer neural network and the second layer neural network may be the same.

Of course, in practical applications, other algorithms can also be used to extract feature data, such as the keras algorithm.

Step 103, using a data detection algorithm to process the feature data to obtain a processing result.

In the embodiment of the present application, the data detection algorithm includes: VAE. Specifically, in this step, the feature data is compared with the third network delay data to obtain a comparison result.

VAE algorithms can identify and repair structural defects in images, such as scratches and dust spots, as well as unstructured defects such as noise and blur. As a generative model, VAE's basic idea is to transform real samples into ideal data distributions through the encoder network, and then pass these data distributions to the decoder network to obtain generated samples. By comparing the generated samples with the real samples, the gap between the two can be obtained, so as to continue training or repair the autoencoder model.

In the embodiment of the present application, the VAE algorithm is applied to the process of detecting sudden changes in network delay. In this process, the VAE algorithm will automatically learn traffic behavior patterns and fluctuation patterns in various forms, and compare the original data (the aforementioned feature data) with the recovered data (data obtained by using the VAE algorithm) through the encoding and decoding process, and then Judging the time delay mutation value in the first network time delay data input in real time, and obtaining fast and accurate detection results. Specifically, here, through the VAE algorithm, the feature data is passed through the encoder to generate intermediate variables, and then the intermediate variables are passed through the decoder to obtain the repaired data. Wherein, the intermediate variables may also be called hidden variables or sampling variables.

Of course, in practical applications, other data detection algorithms can also be used to process feature data, such as the rima algorithm. However, by comparison, when VAE is used, the detection accuracy of network microbursts is higher.

Step 104. Determine whether a network microburst occurs according to the processing result.

If the VAE algorithm is adopted as the data detection algorithm in step 103, then, when the comparison result indicates that the feature data is different from the third network delay data, it is determined that a network microburst occurs; otherwise, it can be determined that no network microburst occurs. mutation.

In the embodiment of the present application, feature extraction algorithm is used to extract the feature data of the acquired first network delay data, and the feature data is detected by data detection algorithm to obtain the processing result. It can be seen from this that in the embodiment of the present application, the feature data is processed through the data detection algorithm, and it is not necessary to set a fixed threshold to judge the network microburst as in the prior art, thereby avoiding network microbursts. Misjudgment of micro-bursts can improve the detection accuracy of network micro-bursts.

Referring to Fig. 2, Fig. 2 is a flow chart of the information detection method provided by the embodiment of the present application, as shown in Fig. 2, including the following steps:

Step 201. Acquire first network delay data.

Step 202. Extract feature data of the first network delay data by using a feature extraction algorithm.

Step 203, using a data detection algorithm to process the feature data to obtain a processing result.

Wherein, for the process from step 201 to step 203, reference may be made to the description of steps 101 to 103 above.

In the above process, LSMT is used to extract the features of the original network delay data, and the original time series data is converted into multi-dimensional data that can be directly imported by VAE, so as to improve the efficiency and accuracy of VAE detection.

Specifically, here, through the VAE algorithm, the feature data is passed through the encoder to generate intermediate variables, and then the intermediate variables are passed through the decoder to obtain the repaired data.

Step 204, using the intermediate variable of VAE to obtain the first score value, wherein the intermediate variable is generated during the process of detecting the feature data by using VAE.

Here, the logarithm operation is mainly performed on the intermediate variable to obtain the first score value. The first fractional value is used to measure the delay value.

Step 205, using the SVR to detect the intermediate variable to obtain a second score value.

In order to further improve the detection accuracy, after using VAE to detect the characteristic data, use SVR to check the intermediate variables, filter out possible noise, and lock the real delay mutation value.

SVR can be used for linear/non-linear classification as well as regression. Different functions can be realized according to the input data. By seeking the minimum structural risk to improve the generalization ability of the learning machine, the empirical risk and the confidence range are minimized, so as to achieve the purpose of obtaining good statistical laws even when the statistical sample size is small. In layman's terms, it is a two-class classification model, and its basic model is defined as a linear classifier with the largest interval in the feature space, that is, the learning strategy of the support vector machine is to maximize the interval, which can eventually be transformed into a convex quadratic Solutions to planning problems.

Specifically, here, the logarithm of the output of the SVR may also be taken to obtain the second score value. The method of taking logarithm to obtain the first score value is the same as the method of delogarithm to obtain the second score value.

Step 206: Determine whether a network microburst occurs according to the processing result.

When the first score value is greater than the second score value, it is determined that a network microburst occurs. Otherwise, it can be regarded as a normal delay value.

In practical applications, the detected delay mutation value not only includes the delay value, but also includes the number of mutation values, so that the duration of the delay mutation can be calculated according to the corresponding occurrence time of each delay value .

On the basis of the above embodiments, before step 205, the SVR model may also be trained using historical score values and historical intermediate variables. Wherein, the historical score value is obtained by using VAE to detect the characteristic data of the historical network delay data, and the historical intermediate variable is generated during the process of using the VAE to detect the characteristic data of the historical network delay data. That is, the historical score value and historical intermediate variable refer to the first score value and intermediate variable obtained through VAE in the information detection method that has been performed before. Specifically, during the training process, the historical intermediate variable is used as a feature, and the historical score value is used as a label to train the SVR model. Wherein, the historical intermediate variable and the historical score value may be acquired during the process of using VAE to determine the occurrence of a network microburst.

In the embodiment of the present application, feature extraction algorithm is used to extract the feature data of the acquired first network delay data, and the feature data is detected by data detection algorithm to obtain the processing result. It can be seen from this that in the embodiment of the present application, the feature data is processed through the data detection algorithm, and it is not necessary to set a fixed threshold to judge the network microburst as in the prior art, thereby avoiding network microbursts. Misjudgment of micro-bursts can improve the detection accuracy of network micro-bursts.

Referring to Fig. 3, Fig. 3 is a flow chart of the information detection method provided by the embodiment of the present application, as shown in Fig. 3, including the following steps:

Step 301. Obtain original time delay data.

Among them, the original delay data and the like can be obtained through the network delay monitoring device.

Step 302: Process the original time-delay data through the double-layer network of LSMT to obtain feature data.

Step 303, input feature data to VAE for processing.

Firstly, it is encoded by the encoding layer of VAE to obtain hidden variables. The hidden variables are then decoded by the decoding layer to obtain the restored data. By calculating the hidden variables, the calculated value of VAE is obtained.

Step 304, use SVR to calculate the hidden variable, and obtain the SVR calculation value.

Step 305, comparing the VAE calculation with the SVR calculation.

When the VAE calculation value is greater than the SVR calculation value, it is determined that the network micro-mutation occurs; otherwise, it can be determined that the network micro-mutation does not occur.

Step 306, display the result.

Wherein, the result may include whether a network micro-mutation occurs, the magnitude of the delay value, the number of mutation values, and so on.

In the above embodiments, the LSTM algorithm is used to preprocess the original network delay data, the VAE algorithm is used to detect the delay mutation value, and the SVR is used to verify the detection results of the VAE detection algorithm. Accurate identification of delay normal value and delay mutation value, so as to realize the detection of network micro-burst occurrence time and duration. Through the above scheme, it is not necessary to judge the network micro-burst by setting a fixed threshold as in the prior art, so that the delay mutation caused by the network micro-burst can be accurately detected, and the delay abnormality caused by other network changes can be avoided. Judgment (such as abnormal time delay caused by network path switching, slow network performance degradation, etc.), improves the accuracy of network microburst detection, and the application is simpler. At the same time, the solutions in the embodiments of the present application are applicable to the detection of network microburst delay mutations in different network scales and different network scenarios.

The embodiment of the present application also provides an information detection device. As shown in Figure 4, the information detection device 400 includes:

The first acquisition module 401 is used to obtain the first network delay data; the first extraction module 402 uses a feature extraction algorithm to extract the feature data of the first network delay data; the first processing module 403 is used to use the data The detection algorithm processes the feature data to obtain a processing result; the first determining module 404 is configured to determine whether a network microburst occurs according to the processing result.

Wherein, the feature extraction algorithm includes LSTM; the first extraction module includes:

A preprocessing submodule, configured to use LSTM to preprocess the first network delay data to obtain second network delay data;

The feature extraction submodule is configured to use the LSTM to perform feature extraction on the second network delay data to obtain the feature data.

Wherein, the feature extraction submodule includes:

The first processing unit is configured to use the first layer neural network of the LSTM to perform pooling processing on the delay data of the second network to obtain intermediate data;

The second processing unit is configured to use the second layer neural network of the LSTM to perform pooling processing on the intermediate data to obtain the feature data.

Wherein, the data detection algorithm includes: VAE; the first processing module is configured to compare the feature data with the third network delay data to obtain a comparison result;

The first determining module is configured to determine that a network microburst occurs when the comparison result indicates that the feature data is different from the third network delay data.

Wherein, the data detection algorithm includes: VAE; the device also includes:

The second acquisition module is used to obtain the first score value by using the intermediate variable of VAE, wherein the intermediate variable is generated during the process of detecting the characteristic data by using VAE

The third acquisition module is used to detect the intermediate variable by using support vector regression SVR to obtain the second score value;

The first determining module is configured to determine that a network microburst occurs when the first score value is greater than the second score value.

Wherein, the device also includes:

The training module is used to train the SVR model by utilizing historical score values and historical intermediate variables;

Wherein, the historical score value is obtained by using VAE to detect the characteristic data of the historical network delay data, and the historical intermediate variable is generated during the process of using the VAE to detect the characteristic data of the historical network delay data.

The device provided in the embodiment of the present application can execute the above-mentioned method embodiment, and its implementation principle and technical effect are similar, and will not be repeated here in this embodiment.

It should be noted that the division of units in the embodiment of the present application is schematic, and is only a logical function division, and there may be another division manner in actual implementation. In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is implemented in the form of a software function unit and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other various media that can store program codes. .

The embodiment of the present application also provides an information detection device. As shown in FIG. 5 , the information detection device in the embodiment of the present application includes: a processor 501 and a transceiver 502 . Wherein, the processor 501 is used for:

Obtain the first network delay data;

Using a feature extraction algorithm to extract feature data of the first network delay data;

Using a data detection algorithm to process the feature data to obtain a processing result;

According to the processing result, it is determined whether a network microburst occurs.

Wherein, the feature extraction algorithm includes LSTM; the processor is used for:

Using LSTM to preprocess the first network delay data to obtain second network delay data;

Using the LSTM to perform feature extraction on the second network delay data to obtain the feature data.

Wherein, the processor 501 is configured to:

Using the first layer neural network of the LSTM to perform pooling processing on the second network delay data to obtain intermediate data;

The second-layer neural network of the LSTM is used to perform pooling processing on the intermediate data to obtain the feature data.

Wherein, the data detection algorithm includes: VAE; the processor 501 is configured to:

Using VAE to detect the feature data to obtain third network delay data;

Comparing the feature data with the third network delay data to obtain a comparison result;

The determining whether a network microburst occurs according to the processing result includes:

When the comparison result indicates that the feature data is different from the third network delay data, it is determined that a network microburst occurs.

Wherein the data detection algorithm includes: VAE; the processor 501 is configured to:

Using an intermediate variable of VAE to obtain a first score value, wherein the intermediate variable is generated during the process of detecting the feature data using VAE;

Using SVR to detect the intermediate variable to obtain a second score value;

The determining whether a network microburst occurs according to the processing result includes:

When the first score value is greater than the second score value, it is determined that a network microburst occurs.

Wherein, the processor is used for:

Use historical score values and historical intermediate variables to train the SVR model;

Wherein, the historical score value is obtained by using VAE to detect the characteristic data of the historical network delay data, and the historical intermediate variable is generated during the process of using the VAE to detect the characteristic data of the historical network delay data.

The device provided in the embodiment of the present application can execute the above-mentioned method embodiment, and its implementation principle and technical effect are similar, and will not be repeated here in this embodiment.

The embodiment of the present application also provides an electronic device, including: a memory, a processor, and a program stored in the memory and operable on the processor, and the processor implements the above-mentioned information detection method when executing the program A step of.

The embodiment of the present application also provides a readable storage medium, on which a program is stored. When the program is executed by a processor, each process of the above-mentioned information detection method embodiment can be achieved, and the same technical effect can be achieved. To avoid Repeat, no more details here. Wherein, the readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor memory (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid state hard drive (SSD)), etc.

It should be noted that, in this document, the term "comprising", "comprising" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, It also includes other elements not expressly listed, or elements inherent in the process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising that element.

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

The embodiments of the present application have been described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementations. The above-mentioned specific implementations are only illustrative and not restrictive. Those of ordinary skill in the art will Under the inspiration of this application, without departing from the purpose of this application and the scope of protection of the claims, many forms can also be made, all of which belong to the protection of this application.

Claims (10)

1. An information detection method, characterized in that, comprising: Obtain the first network delay data; Using a feature extraction algorithm to extract feature data of the first network delay data; Using a data detection algorithm to process the feature data to obtain a processing result; According to the processing result, it is determined whether a network microburst occurs. 2. The method according to claim 1, wherein the feature extraction algorithm comprises long short-term memory (LSTM); The extraction of feature data of the first network delay data using a feature extraction algorithm includes: Using LSTM to preprocess the first network delay data to obtain second network delay data; Using the LSTM to perform feature extraction on the second network delay data to obtain the feature data. 3. The method according to claim 2, wherein said utilizing said LSTM to perform feature extraction on said second network delay data to obtain said feature data comprises: Using the first layer neural network of the LSTM to perform pooling processing on the second network delay data to obtain intermediate data; The second-layer neural network of the LSTM is used to perform pooling processing on the intermediate data to obtain the feature data. 4. The method according to claim 1, wherein the data detection algorithm comprises: variational autoencoder VAE; the processing of the feature data using the data detection algorithm comprises: Using VAE to detect the feature data to obtain third network delay data; Comparing the feature data with the third network delay data to obtain a comparison result; The determining whether a network microburst occurs according to the processing result includes: When the comparison result indicates that the feature data is different from the third network delay data, it is determined that a network microburst occurs. 5. The method according to claim 1, wherein the data detection algorithm comprises: VAE; after the feature data is processed using the data detection algorithm and the processing result is obtained, the method further comprises: Using an intermediate variable of VAE to obtain a first score value, wherein the intermediate variable is generated during the process of detecting the feature data using VAE; Using support vector regression SVR to detect the intermediate variable to obtain a second score value; The determining whether a network microburst occurs according to the processing result includes: When the first score value is greater than the second score value, it is determined that a network microburst occurs. 6. The method according to claim 5, further comprising: Use historical score values and historical intermediate variables to train the SVR model; Wherein, the historical score value is obtained by using VAE to detect the characteristic data of the historical network delay data, and the historical intermediate variable is generated during the process of using the VAE to detect the characteristic data of the historical network delay data. 7. An information detection device, characterized in that it comprises: The first obtaining module is used to obtain the first network delay data; The first extraction module, the module uses a feature extraction algorithm to extract the feature data of the first network delay data; The first processing module is used to process the feature data by using a data detection algorithm to obtain a processing result; The first determining module is configured to determine whether a network microburst occurs according to the processing result. 8. An information detection device, comprising: a processor and a transceiver; Wherein, the processor is used for: Obtain the first network delay data; Using a feature extraction algorithm to extract feature data of the first network delay data; Using a data detection algorithm to process the feature data to obtain a processing result; According to the processing result, it is determined whether a network microburst occurs. 9. An electronic device, comprising: a memory, a processor, and a program stored on the memory and operable on the processor; it is characterized in that, The processor is configured to read the program in the memory to implement the steps in the information detection method according to any one of claims 1 to 6. 10. A readable storage medium for storing a program, wherein when the program is executed by a processor, the steps in the information detection method according to any one of claims 1 to 6 are realized.

Images