A method and system for business indicator situation awareness of enterprise network resources

Abstract

The present invention relates to the field of network communication technology, and in particular to a method and system for business indicator situation awareness of enterprise network resources. The present invention determines the eligibility of a single complete enterprise network resource data according to a similarity coefficient, screens out a single complete enterprise network resource data with the largest similarity coefficient, and fills in missing characteristic variables in the single complete enterprise network resource data with corresponding characteristic variables in the single complete enterprise network resource data, and pre-processes the enterprise network resource data; determines a business indicator evaluation coefficient based on the enterprise network resource data obtained by predicting future enterprise network resource data according to a time series prediction model, and determines an early warning for the enterprise network resources. The present invention overcomes the problem in the prior art that the integrity of enterprise network resource data is poor and the accuracy of supplementing missing data is low, resulting in poor prediction accuracy of the prediction model, thereby resulting in poor accuracy of business indicator situation awareness.

Description

Technical Field

[0001] The present invention relates to the field of network communication technology, and in particular to a method and system for sensing business indicator situation of enterprise network resources. Background Art

[0002] In today's information technology era, the management and optimization of enterprise network resources are becoming increasingly important. Enterprise network resources, including bandwidth, servers, storage devices, etc., are the basis for the normal operation of enterprise information systems. With the increase in business volume and the development of network technology, enterprise network resources are facing increasing pressure. How to effectively manage and optimize these resources and ensure the stability and efficiency of enterprise information systems has become an important issue in enterprise information technology management.

[0003] Traditional enterprise network resource management methods mainly rely on manual experience and simple monitoring tools. This method is not only inefficient, but also difficult to adapt to the rapidly changing network environment and business needs. For example, when network traffic suddenly increases and causes network congestion, traditional methods often fail to detect and solve the problem in time, thus affecting the normal operation of enterprise business.

[0004] In order to solve the above problems, in recent years, with the development of big data and artificial intelligence technology, business indicator situation awareness methods have emerged. This method collects and analyzes a large amount of network resource data, uses advanced data analysis and machine learning algorithms, and monitors and predicts the use of enterprise network resources and business indicators in real time, thereby realizing intelligent management and optimization of enterprise network resources. Through this method, enterprises can promptly discover and warn of problems that may affect network performance and business operations, take measures to make adjustments in advance, and ensure the efficient use of network resources and the stable operation of enterprise business.

[0005] Chinese patent application publication number: CN109766695A discloses a network security situation awareness method and system based on fusion decision-making, including collecting sF low data, NetF low data, SNMP data and log data from a monitored network; preprocessing the collected sF low data, NetF low data, SNMP data and log data; performing traffic analysis and device performance analysis on the preprocessed data, and fusing them through a fusion algorithm to obtain the security situation of the monitored network.

[0006] It can be seen that the existing technology has the following problems: due to the poor integrity of enterprise network resource data and the low accuracy of supplementing missing data, the prediction accuracy of the prediction model is poor, which leads to poor accuracy of business indicator situation awareness. Summary of the invention

[0007] To this end, the present invention provides a method and system for business indicator situation awareness of enterprise network resources, so as to overcome the problem in the prior art that the integrity of enterprise network resource data is poor and the accuracy of supplementing missing data is low, resulting in poor prediction accuracy of the prediction model, thereby resulting in poor accuracy of business indicator situation awareness.

[0008] To achieve the above object, on the one hand, the present invention provides a method for business indicator situation awareness of enterprise network resources, comprising:

[0009] Obtain enterprise network resource data within a preset historical period;

[0010] Determining whether there is a missing feature variable in the single enterprise network resource data based on a first number of feature variables in the single enterprise network resource data;

[0011] Determine the eligibility of the single complete enterprise network resource data according to the similarity coefficient between the single enterprise network resource data and the single complete enterprise network resource data, screen out the single complete enterprise network resource data with the largest similarity coefficient among the qualified complete enterprise network resource data, and fill the missing characteristic variables in the single enterprise network resource data with the corresponding characteristic variables in the single complete enterprise network resource data; when there is no qualified complete enterprise network resource data, delete the single enterprise network resource data;

[0012] Or, determining whether there is missing data between the single enterprise network resource data and the previous enterprise network resource data according to the time difference between the timestamp of the single enterprise network resource data and the timestamp of the previous enterprise network resource data;

[0013] Determining preprocessing of the enterprise network resource data according to a second amount of missing data of the enterprise network resource data within a preset historical time period;

[0014] Training enterprise network resource data on a time series prediction model, and predicting future enterprise network resource data based on the trained time series prediction model;

[0015] A business indicator evaluation coefficient is determined according to the predicted enterprise network resource data, and an early warning for the enterprise network resources is determined according to the business indicator evaluation coefficient.

[0016] Further, the determining whether there is a missing characteristic variable in the single enterprise network resource data includes determining whether there is a missing characteristic variable in the single enterprise network resource data under the condition that a first number of characteristic variables in the single enterprise network resource data is less than or equal to a first preset number.

[0017] Furthermore, the similarity coefficient between the single enterprise network resource data and the single complete enterprise network resource data is calculated according to the following formula:

[0018]

[0019] Among them, J represents the similarity coefficient, Ai represents the i-th characteristic variable in the single enterprise network resource data, Ai0 represents the i-th characteristic variable in a single complete enterprise network resource data, and n represents the number of characteristic variables in the single enterprise network resource data.

[0020] Further, the determining the eligibility of the single complete enterprise network resource data includes determining that the single complete enterprise network resource data is qualified under the condition that the similarity coefficient is greater than a preset similarity coefficient.

[0021] Further, the determining whether there is missing data between the single enterprise network resource data and the previous enterprise network resource data includes determining whether there is missing data between the single enterprise network resource data and the previous enterprise network resource data under the condition that the time difference is greater than a preset time difference.

[0022] Further, the determining to pre-process the enterprise network resource data includes determining to pre-process the enterprise network resource data under the condition that the second number is greater than a second preset number.

[0023] Furthermore, the preprocessing of the enterprise network resource data includes averaging the first K enterprise network resource data and the last K enterprise network resource data of a single missing data in the enterprise network resource data under the condition that the relative difference is less than or equal to a preset relative difference to fill the single missing data.

[0024] Further, the preprocessing of the enterprise network resource data includes adjusting the relative difference according to the adjustment coefficient under the condition that the relative difference is greater than the preset relative difference. The preset historical duration is adjusted, wherein ΔL represents the relative difference, and ΔL0 represents the preset relative difference.

[0025] Furthermore, the business indicator evaluation coefficient is calculated according to the following formula:

[0026]

[0027] Wherein, Z represents the business indicator evaluation coefficient, Bv1 represents the bandwidth utilization rate in the v-th enterprise network resource data within the preset future time t, Bv2 represents the TCP proportion in the v-th enterprise network resource data within the preset future time t, Bv3 represents the UDP proportion in the v-th enterprise network resource data within the preset future time t, Bv4 represents the HTTP proportion in the v-th enterprise network resource data within the preset future time t, Bv5 represents the FTP proportion in the v-th enterprise network resource data within the preset future time t, and N represents the number of enterprise network resource data within the preset future time t;

[0028] The determining of the early warning for the enterprise network resources includes performing an early warning for the enterprise network resources under the condition that the business indicator evaluation coefficient is greater than a preset business indicator evaluation coefficient.

[0029] On the other hand, the present invention also provides a system for using a business indicator situation awareness method of enterprise network resources, comprising:

[0030] A data acquisition unit, which is used to acquire enterprise network resource data within a preset historical period;

[0031] A first analysis unit, connected to the data acquisition unit, for determining whether there is a missing characteristic variable in the single enterprise network resource data according to a first number of characteristic variables in the single enterprise network resource data;

[0032] a second analysis unit, which is connected to the data acquisition unit and the first analysis unit respectively, and is used to determine the eligibility of the single complete enterprise network resource data according to the similarity coefficient between the single enterprise network resource data and the single complete enterprise network resource data, screen out the single complete enterprise network resource data with the largest similarity coefficient among the qualified complete enterprise network resource data, and fill the missing characteristic variables in the single enterprise network resource data with the corresponding characteristic variables in the single complete enterprise network resource data, and delete the single enterprise network resource data if there is no qualified complete enterprise network resource data;

[0033] A third analysis unit, which is connected to the data acquisition unit, the first analysis unit and the second analysis unit respectively, and is used to determine whether there is missing data between the single enterprise network resource data and the previous enterprise network resource data according to the time difference between the timestamp of the single enterprise network resource data and the timestamp of the previous enterprise network resource data;

[0034] Determining preprocessing of the enterprise network resource data according to a second amount of missing data of the enterprise network resource data within a preset historical time period;

[0035] an adjustment unit, connected to the third analysis unit, for adjusting the preset historical duration;

[0036] A model training unit, connected to the third analysis unit, for training the enterprise network resource data on the time series prediction model, and predicting the future enterprise network resource data according to the trained time series prediction model;

[0037] An early warning unit is connected to the model training unit and is used to determine a business indicator evaluation coefficient based on the predicted enterprise network resource data, and to determine an early warning for the enterprise network resources based on the business indicator evaluation coefficient.

[0038] Compared with the prior art, the beneficial effect of the present invention lies in that the present invention automatically detects whether there are missing feature variables in the enterprise network resource data, thereby reducing the workload of manual inspection and improving the efficiency of data preprocessing. By timely discovering and processing missing feature variables to ensure the data integrity of subsequent analysis and model training, the accuracy and reliability of model prediction can be improved, and the monitoring and early warning of enterprise network resources can be ensured to be more accurate, thereby improving network efficiency and the stability of business operations.

[0039] Furthermore, the present invention calculates the similarity coefficient between a single enterprise network resource data and a single complete enterprise network resource data through a formula, identifies and screens out qualified data, and supplements the missing characteristic variables with the single complete enterprise network resource data that is most similar to the single enterprise network resource data, thereby improving the efficiency and accuracy of data processing and avoiding the low prediction accuracy of the time series prediction model due to poor data quality.

[0040] Furthermore, the present invention can identify missing data in a data set, thereby improving the efficiency of data preprocessing. By timely discovering missing data, corresponding remedial measures can be taken to ensure the integrity of the data set. Missing data will have a negative impact on data analysis and model training. By ensuring the integrity of the data, the accuracy and reliability of the time series prediction model can be improved, thereby providing enterprises with more accurate business insights and decision support.

[0041] Furthermore, the present invention determines whether to preprocess the enterprise network resource data based on a comparison result of the second number of missing data of the enterprise network resource data within a preset historical period and a second preset number, so as to perform preprocessing when there is a large amount of missing data to significantly improve the integrity and quality of the data, thereby improving the prediction accuracy of the model and providing a more accurate basis for subsequent data analysis and decision-making.

[0042] Furthermore, the present invention flexibly selects the most appropriate preprocessing method according to the actual situation of data missing, ensuring that the preprocessing measures can not only effectively deal with the problem of data missing, but also avoid waste of resources caused by excessive processing, and selects the preprocessing method that is most suitable for the current data missing situation to more effectively restore the integrity and accuracy of the data, thereby improving the overall data quality and providing a more accurate and reliable data foundation for subsequent data analysis and model training, thereby improving the accuracy and credibility of business indicator situation awareness.

[0043] Furthermore, the present invention adjusts the preset historical duration by calculating the adjustment coefficient according to the formula to accurately extend the preset historical duration, obtains more enterprise network resource data to provide reliable data input for the time series prediction model, thereby improving the prediction accuracy of the model.

[0044] Furthermore, the present invention determines whether to issue an enterprise network resource warning based on the comparison result of the business indicator evaluation coefficient with the preset business indicator evaluation coefficient, timely discovers the bottlenecks faced by the enterprise network resources, and thus takes preventive or mitigation measures. By monitoring and warning the key business indicator evaluation coefficients, it helps the enterprise to more effectively manage and optimize the allocation and use of network resources, thereby improving the overall utilization efficiency of network resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flow chart of a method for business indicator situation awareness of enterprise network resources according to an embodiment of the present invention;

[0046] Figure 2 A schematic diagram of the structure of a system for applying a method for business indicator situation awareness of enterprise network resources according to an embodiment of the present invention;

[0047] Figure 3 The present invention is a logical block diagram of a method for business indicator situation awareness of enterprise network resources according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0049] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0050] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0051] See also Figure 1 , 3 As shown, Figure 1 It is a flow chart of a method for business indicator situation awareness of enterprise network resources according to an embodiment of the present invention; Figure 3 The present invention is a logical block diagram of a method for business indicator situation awareness of enterprise network resources according to an embodiment of the present invention.

[0052] The method for business indicator situation awareness of enterprise network resources according to an embodiment of the present invention includes:

[0053] Step S1, obtaining enterprise network resource data within a preset historical time period T;

[0054] Step S2, determining whether there is a missing characteristic variable in the single enterprise network resource data according to the first number D of characteristic variables in the single enterprise network resource data;

[0055] Step S3, determining the eligibility of the single complete enterprise network resource data according to the similarity coefficient J between the single enterprise network resource data and the single complete enterprise network resource data, screening out the single complete enterprise network resource data with the largest similarity coefficient J among the qualified complete enterprise network resource data, and filling the missing characteristic variables in the single enterprise network resource data with the corresponding characteristic variables in the single complete enterprise network resource data, and deleting the single enterprise network resource data when there is no qualified complete enterprise network resource data;

[0056] Or, determining whether there is missing data between the single enterprise network resource data and the previous enterprise network resource data according to the time difference C between the timestamp of the single enterprise network resource data and the timestamp of the previous enterprise network resource data;

[0057] Step S4, determining preprocessing of the enterprise network resource data according to a second number L of missing data of the enterprise network resource data within a preset historical time length T;

[0058] Step S5, training the enterprise network resource data on the time series prediction model, and predicting the future enterprise network resource data according to the trained time series prediction model;

[0059] Step S6, determining a business indicator evaluation coefficient according to the predicted enterprise network resource data, and determining whether to perform an enterprise network resource warning according to the business indicator evaluation coefficient.

[0060] In an embodiment of the present invention, a single enterprise network resource data includes several characteristic variables with timestamps, and the several characteristic variables are the bandwidth utilization, TCP proportion, UDP proportion, HTTP proportion, and FTP proportion of the enterprise network. The time series prediction model is preferably an LSTM model, and the preset historical duration T is 30 days. Technical personnel in this field can adjust the preset historical duration T according to specific circumstances.

[0061] Specifically, it is determined whether there is a missing characteristic variable in the single enterprise network resource data according to the comparison result of the first number D of characteristic variables in the single enterprise network resource data and the first preset number D0;

[0062] When D<D0, it is determined that there is a missing characteristic variable in the single enterprise network resource data;

[0063] When D=D0, it is determined that there is no missing feature variable in the single enterprise network resource data;

[0064] In the embodiment of the present invention, the first preset number D0 is the number of all characteristic variables in a single complete enterprise network resource data.

[0065] Specifically, the present invention automatically detects whether there are missing feature variables in the enterprise network resource data, thereby reducing the workload of manual inspection and improving the efficiency of data preprocessing. By timely discovering and processing missing feature variables to ensure the data integrity of subsequent analysis and model training, the accuracy and reliability of model predictions can be improved, and more accurate monitoring and early warning of enterprise network resources can be ensured, thereby improving network efficiency and the stability of business operations.

[0066] Specifically, when there is a missing characteristic variable in a single enterprise network resource data, the similarity coefficient J between the single enterprise network resource data and a single complete enterprise network resource data is calculated according to the following formula, and is set:

[0067]

[0068] Among them, Ai represents the i-th characteristic variable in the single enterprise network resource data, Ai0 represents the i-th characteristic variable in a single complete enterprise network resource data, and n represents the number of characteristic variables in the single enterprise network resource data.

[0069] Specifically, when the similarity coefficient J is calculated, whether the single complete enterprise network resource data is qualified is determined according to the comparison result between the similarity coefficient J and the preset similarity coefficient J0;

[0070] When J≤J0, it is determined that the single complete enterprise network resource data is unqualified;

[0071] When J>J0, it is determined that the single complete enterprise network resource data is qualified;

[0072] In the embodiment of the present invention, the preset similarity coefficient J0 is set to 0.8, and those skilled in the art may adjust the preset similarity coefficient J0 according to specific circumstances.

[0073] Specifically, the present invention calculates the similarity coefficient between a single enterprise network resource data and a single complete enterprise network resource data through a formula, identifies and screens out qualified data, and supplements the missing characteristic variables with the single complete enterprise network resource data that is most similar to the single enterprise network resource data, thereby improving the efficiency and accuracy of data processing and avoiding the low prediction accuracy of the time series prediction model due to poor data quality.

[0074] Specifically, when there is no missing characteristic variable in all enterprise network resource data, it is determined whether there is missing data between a single enterprise network resource data and the previous enterprise network resource data according to the comparison result of the time difference C and the preset time difference C0;

[0075] When C≤C0, it is determined that there is no missing data between a single enterprise network resource data and the previous enterprise network resource data;

[0076] When C>C0, it is determined that there is missing data between the single enterprise network resource data and the previous enterprise network resource data;

[0077] In the embodiment of the present invention, the preset time difference C0 is obtained by taking the average value of the time difference between the timestamps of all enterprise network resource data within the preset historical time period T. Those skilled in the art may adjust the preset time difference C0 according to specific circumstances.

[0078] Specifically, the present invention can identify missing data in a data set, improve the efficiency of data preprocessing, and take corresponding remedial measures by timely discovering missing data. Missing data will have a negative impact on data analysis and model training. By ensuring the integrity of data, the accuracy and reliability of the time series prediction model can be improved, providing enterprises with more accurate business insights and decision support.

[0079] Specifically, when the missing data of the enterprise network resource data is determined to be complete, it is determined whether to pre-process the enterprise network resource data according to the comparison result of the second number L of missing data of the enterprise network resource data within the preset historical time length T and the second preset number L0;

[0080] When L≤L0, it is determined that the enterprise network resource data is not preprocessed;

[0081] When L>L0, it is determined to pre-process the enterprise network resource data;

[0082] In the embodiment of the present invention, the second preset number L0 is set to 10% of the second number of enterprise network resource data within the preset historical time period T. Those skilled in the art may adjust the second preset number L0 according to specific circumstances.

[0083] Specifically, the present invention determines whether to preprocess the enterprise network resource data based on the comparison result of the second number of missing data of the enterprise network resource data within a preset historical period and the second preset number, so as to perform preprocessing when there is a large amount of missing data to significantly improve the integrity and quality of the data, thereby improving the prediction accuracy of the model and providing a more accurate basis for subsequent data analysis and decision-making.

[0084] Specifically, when it is determined to pre-process the enterprise network resource data, a relative difference △L between the second number L and the second preset number L0 is calculated, and a pre-processing method for pre-processing the enterprise network resource data is determined according to a comparison result of the relative difference △L and the preset relative difference △L0, and △L=(L-L0) / L0 is set;

[0085] When △L≤△L0, it is determined to pre-process the enterprise network resource data in a first pre-processing manner;

[0086] When △L>△L0, it is determined to pre-process the enterprise network resource data in the second pre-processing manner;

[0087] The first preprocessing method is to take the average of the first K enterprise network resource data and the last K enterprise network resource data of a single missing data in the enterprise network resource data to fill the single missing data;

[0088] The second preprocessing method is to adjust the preset historical time length T.

[0089] In the embodiment of the present invention, the preset relative difference △L0 is 2, and the preset relative difference △L0 is obtained when the second number L is 30%. The value of K is preferably 3. Those skilled in the art can adjust the preset relative difference △L0 and the value of K according to the specific situation.

[0090] Specifically, the present invention flexibly selects the most appropriate preprocessing method according to the actual situation of data missing, ensuring that the preprocessing measures can not only effectively deal with the problem of data missing, but also avoid waste of resources caused by excessive processing. It selects the preprocessing method that best suits the current data missing situation to more effectively restore the integrity and accuracy of the data, and improves the overall data quality to provide a more accurate and reliable data foundation for subsequent data analysis and model training, thereby improving the accuracy and credibility of business indicator situation awareness.

[0091] Specifically, when it is determined to pre-process the enterprise network resource data in the second pre-processing manner, the preset historical duration T is adjusted according to the following adjustment coefficient X, and is set as:

[0092]

[0093] The adjusted preset historical time length is set to TR=T×X.

[0094] Specifically, the present invention calculates an adjustment coefficient according to a formula to adjust the preset historical duration to accurately extend the preset historical duration, obtains more enterprise network resource data to provide reliable data input for the time series prediction model, thereby improving the prediction accuracy of the model.

[0095] Specifically, the business indicator evaluation coefficient Z is calculated according to the following formula:

[0096]

[0097] Among them, Bv1 represents the bandwidth utilization rate in the vth enterprise network resource data within the preset future time t, Bv2 represents the TCP proportion in the vth enterprise network resource data within the preset future time t, Bv3 represents the UDP proportion in the vth enterprise network resource data within the preset future time t, Bv4 represents the HTTP proportion in the vth enterprise network resource data within the preset future time t, Bv5 represents the FTP proportion in the vth enterprise network resource data within the preset future time t, and N represents the number of enterprise network resource data within the preset future time t.

[0098] In the embodiment of the present invention, the preset future time length t is set to 24 hours, and those skilled in the art may adjust the preset future time length t according to specific circumstances.

[0099] Specifically, when the business indicator evaluation coefficient Z is calculated, it is determined whether to perform an enterprise network resource warning according to the comparison result between the business indicator evaluation coefficient Z and the preset business indicator evaluation coefficient Z0;

[0100] When Z≤Z0, it is determined that no enterprise network resource warning will be issued;

[0101] When Z>Z0, it is determined to carry out enterprise network resource warning;

[0102] In the embodiment of the present invention, the preset business indicator evaluation coefficient Z0 is set to 80%. Those skilled in the art may adjust the preset business indicator evaluation coefficient Z0 according to specific circumstances.

[0103] Specifically, the present invention determines whether to issue an enterprise network resource warning based on the comparison result of the business indicator evaluation coefficient and the preset business indicator evaluation coefficient, timely discovers the bottlenecks faced by the enterprise network resources, and thus takes preventive or mitigation measures. By monitoring and warning the key business indicator evaluation coefficients, it helps the enterprise to more effectively manage and optimize the allocation and use of network resources, thereby improving the overall utilization efficiency of network resources.

[0104] Please continue reading Figure 2 As shown, Figure 2 A schematic diagram of the structure of a system for applying a method for business indicator situation awareness of enterprise network resources according to an embodiment of the present invention;

[0105] The system of the embodiment of the present invention using the method for business indicator situation awareness of enterprise network resources includes:

[0106] A data acquisition unit, which is used to acquire enterprise network resource data within a preset historical time period T;

[0107] A first analysis unit, connected to the data acquisition unit, for determining whether there is a missing characteristic variable in the single enterprise network resource data according to a first quantity D of characteristic variables in the single enterprise network resource data;

[0108] a second analysis unit, which is connected to the data acquisition unit and the first analysis unit respectively, and is used to determine the eligibility of the single complete enterprise network resource data according to the similarity coefficient J between the single enterprise network resource data and the single complete enterprise network resource data, screen out the single complete enterprise network resource data with the largest similarity coefficient J among the qualified complete enterprise network resource data, and fill the missing characteristic variables in the single enterprise network resource data with the corresponding characteristic variables in the single complete enterprise network resource data, and delete the single enterprise network resource data if there is no qualified complete enterprise network resource data;

[0109] A third analysis unit, which is connected to the data acquisition unit, the first analysis unit and the second analysis unit respectively, and is used to determine whether there is missing data between the single enterprise network resource data and the previous enterprise network resource data according to the time difference C between the timestamp of the single enterprise network resource data and the timestamp of the previous enterprise network resource data;

[0110] Determine preprocessing of the enterprise network resource data according to a second number L of missing data of the enterprise network resource data within a preset historical time length T;

[0111] an adjustment unit, connected to the third analysis unit, for adjusting the preset historical duration T;

[0112] A model training unit, connected to the third analysis unit, for training the enterprise network resource data on the time series prediction model, and predicting the future enterprise network resource data according to the trained time series prediction model;

[0113] An early warning unit is connected to the model training unit and is used to determine a business indicator evaluation coefficient based on the predicted enterprise network resource data, and to determine an early warning for the enterprise network resources based on the business indicator evaluation coefficient.

[0114] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for business indicator situation awareness of enterprise network resources, characterized in that: include: Obtain enterprise network resource data within a preset historical period; Determining whether there is a missing feature variable in the single enterprise network resource data based on a first number of feature variables in the single enterprise network resource data; Determine the eligibility of the single complete enterprise network resource data according to the similarity coefficient between the single enterprise network resource data and the single complete enterprise network resource data, screen out the single complete enterprise network resource data with the largest similarity coefficient among the qualified complete enterprise network resource data, and fill the missing characteristic variables in the single enterprise network resource data with the corresponding characteristic variables in the single complete enterprise network resource data; when there is no qualified complete enterprise network resource data, delete the single enterprise network resource data; When there is no missing feature variable in all enterprise network resource data, determine whether there is missing data between the single enterprise network resource data and the previous enterprise network resource data according to the time difference between the timestamp of the single enterprise network resource data and the timestamp of the previous enterprise network resource data, and determine whether there is missing data between the single enterprise network resource data and the previous enterprise network resource data under the condition that the time difference is greater than a preset time difference; Determine preprocessing of the enterprise network resource data according to a second number of missing data of the enterprise network resource data within a preset historical duration, the preprocessing comprising averaging the first K enterprise network resource data and the last K enterprise network resource data of a single missing data in the enterprise network resource data under the condition that the relative difference is less than or equal to the preset relative difference to fill the single missing data, wherein the relative difference is the relative difference between the second number and a second preset number, and the second preset number is 10% of the second number of the enterprise network resource data within the preset historical duration T; Training enterprise network resource data on a time series prediction model, and predicting future enterprise network resource data based on the trained time series prediction model; Determining a business indicator evaluation coefficient based on the predicted enterprise network resource data, and determining an early warning for the enterprise network resources based on the business indicator evaluation coefficient; The similarity coefficient between the single enterprise network resource data and the single complete enterprise network resource data is calculated according to the following formula: Among them, J represents the similarity coefficient, Ai represents the i-th characteristic variable in the single enterprise network resource data, Ai0 represents the i-th characteristic variable in a single complete enterprise network resource data, and n represents the number of characteristic variables in the single enterprise network resource data.

2. The method for business indicator situation awareness of enterprise network resources according to claim 1, characterized in that: The determining whether there is a missing characteristic variable in the single enterprise network resource data includes determining whether there is a missing characteristic variable in the single enterprise network resource data under the condition that a first number of characteristic variables in the single enterprise network resource data is less than or equal to a first preset number.

3. The method for business indicator situation awareness of enterprise network resources according to claim 1, characterized in that: The determining the eligibility of the single complete enterprise network resource data includes determining that the single complete enterprise network resource data is qualified under the condition that the similarity coefficient is greater than a preset similarity coefficient.

4. The method for business indicator situation awareness of enterprise network resources according to claim 1, characterized in that: The determining to pre-process the enterprise network resource data includes determining to pre-process the enterprise network resource data under the condition that the second number is greater than a second preset number.

5. The method for business indicator situation awareness of enterprise network resources according to claim 1, characterized in that: The preprocessing of the enterprise network resource data includes adjusting the relative difference according to the adjustment coefficient under the condition that the relative difference is greater than the preset relative difference. The preset historical duration is adjusted, wherein ΔL represents the relative difference, and ΔL0 represents the preset relative difference.

6. The method for business indicator situation awareness of enterprise network resources according to claim 5, characterized in that: The business indicator evaluation coefficient is calculated according to the following formula: Wherein, Z represents the business indicator evaluation coefficient, Bv1 represents the bandwidth utilization rate in the v-th enterprise network resource data within the preset future time t, Bv2 represents the TCP proportion in the v-th enterprise network resource data within the preset future time t, Bv3 represents the UDP proportion in the v-th enterprise network resource data within the preset future time t, Bv4 represents the HTTP proportion in the v-th enterprise network resource data within the preset future time t, Bv5 represents the FTP proportion in the v-th enterprise network resource data within the preset future time t, and N represents the number of enterprise network resource data within the preset future time t; The determining of the early warning for the enterprise network resources includes performing an early warning for the enterprise network resources under the condition that the business indicator evaluation coefficient is greater than a preset business indicator evaluation coefficient.

7. A system using the method for business indicator situation awareness of enterprise network resources according to any one of claims 1 to 6, characterized in that: include: A data acquisition unit, which is used to acquire enterprise network resource data within a preset historical period; A first analysis unit, connected to the data acquisition unit, for determining whether there is a missing characteristic variable in the single enterprise network resource data according to a first number of characteristic variables in the single enterprise network resource data; a second analysis unit, which is connected to the data acquisition unit and the first analysis unit respectively, and is used to determine the eligibility of the single complete enterprise network resource data according to the similarity coefficient between the single enterprise network resource data and the single complete enterprise network resource data, screen out the single complete enterprise network resource data with the largest similarity coefficient among the qualified complete enterprise network resource data, and fill the missing characteristic variables in the single enterprise network resource data with the corresponding characteristic variables in the single complete enterprise network resource data, and delete the single enterprise network resource data if there is no qualified complete enterprise network resource data; A third analysis unit, which is connected to the data acquisition unit, the first analysis unit and the second analysis unit respectively, and is used to determine whether there is missing data between the single enterprise network resource data and the previous enterprise network resource data according to the time difference between the timestamp of the single enterprise network resource data and the timestamp of the previous enterprise network resource data; Determining preprocessing of the enterprise network resource data according to a second amount of missing data of the enterprise network resource data within a preset historical time period; an adjustment unit, connected to the third analysis unit, for adjusting the preset historical duration; A model training unit, connected to the third analysis unit, for training the enterprise network resource data on the time series prediction model, and predicting the future enterprise network resource data according to the trained time series prediction model; An early warning unit is connected to the model training unit and is used to determine a business indicator evaluation coefficient based on the predicted enterprise network resource data, and to determine an early warning for the enterprise network resources based on the business indicator evaluation coefficient.

Images