An artificial intelligence-based network device operation data management system and method

Abstract

The application relates to the technical field of data management, in particular to a network equipment operation data management system and method based on artificial intelligence; the system comprises a data acquisition unit, a data processing unit and a data management unit; the application calculates the calling probability of operation data and the mutual information among the operation data, establishes hidden association, forms a first data set and a second data set, breaks the limitation that traditional data management only focuses on data itself and ignores the association among data, constructs a first buffer data set and a second buffer data set according to a buffer preset threshold value, the first data set and the second data set, and updates the buffer data sets in real time, the frequently-used data and the associated data can be stored in a buffer area in advance, when operation data needs to be called, the reaction speed of the data is greatly improved, the data reading delay is reduced, the data usage frequency and the association are fully considered, and the overall performance of the system and the data calling efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of data management technology, specifically to a network device operation data management system and method based on artificial intelligence. Background Technology

[0002] In today's digital age, network technology is developing rapidly, and network equipment is being used more and more widely and deeply in various fields. From enterprise office networks and data centers to the global internet infrastructure, the number and types of network equipment are constantly increasing. They undertake key tasks such as data transmission, storage, and processing, becoming an important cornerstone supporting the information-based operation of modern society.

[0003] Traditional data collection methods often rely on manual configuration and operation, which is inefficient and prone to errors. At the same time, traditional methods usually only perform simple storage and basic statistical analysis on the collected data, which cannot deeply explore the potential value behind the data, make it difficult to discover the complex relationships between data, and have a slow response speed when retrieving running data.

[0004] Based on this, the present invention provides a network device operation data management system and method based on artificial intelligence to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide a network device operation data management system and method based on artificial intelligence. It breaks through the limitation of traditional data management that only focuses on the data itself and ignores the relationship between data. At the same time, it can pre-store frequently used data and related data in a buffer area, which greatly improves the data response speed and reduces the data reading delay when the operation data needs to be retrieved. It fully considers the frequency of data use and the correlation of data, improves the overall performance of the system and the efficiency of data retrieval. In addition, it continuously updates historical operation data and retrieval records, realizing a dynamic data buffer storage effect.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A first aspect of the present invention provides a network device operation data management system based on artificial intelligence, comprising a data acquisition unit, a data processing unit, and a data management unit, wherein:

[0008] The data acquisition unit is used to collect the operating data of the network device and preprocess the collected operating data;

[0009] The data processing unit is used to construct a first data set, and based on the constructed first data set, construct a second data set among the collected running data, and based on the buffer preset threshold, the constructed first data set and the second data set, construct a first buffer dataset and a second buffer dataset respectively, and update the first buffer dataset and the second buffer dataset. The data processing unit is connected to the data acquisition unit.

[0010] The data management unit is used to display the received information and preset buffer thresholds. The data management unit is connected to both the data acquisition unit and the data processing unit.

[0011] The present invention is further configured such that: the data acquisition unit includes a data acquisition module, a data identification module, and a first communication module, wherein:

[0012] The data acquisition module is used to collect operational data information of network devices;

[0013] The data identification module is used to assign type and time identifiers to the collected operational data, and the data identification module is connected to the data acquisition module.

[0014] The first communication module is used to realize information interaction between the data acquisition unit, the data processing unit, and the data management unit.

[0015] The present invention is further configured such that: the data processing unit includes a second communication module, a first data module, a second data module, a buffer data module, a data update module, and a database module, wherein:

[0016] The second communication module is used to realize information interaction between the data processing unit, the data acquisition unit, and the data management unit;

[0017] The first data module is used to construct a first data set. The first data module is connected to the second communication module. The process of constructing the first data set is as follows:

[0018] Obtain historical running data retrieved within a historical data period T, along with the corresponding number of retrievals, and calculate the retrieval probability for each set of running data. In the formula, C i The number of times the i-th type of running data is retrieved within period T;

[0019] Calculate the mutual information between various running data. In the formula, X and Y are the running data D. i and D j The set of possible values ​​for p(x, y) is D. i =x and D j=The joint probability of y, p(x) and p(y) are respectively D i =x and D j =marginal probability of y;

[0020] The historical running data is sorted according to the number of times it is retrieved, and implicit relationships between the sorted historical running data are established based on the mutual information between the running data to obtain the first data set.

[0021] The calculation process for the joint probability is as follows:

[0022] Create an n×n matrix M to record the number of co-occurrences among different running data;

[0023] Iterate through all data retrieval records. For each retrieval operation, if the running data D is retrieved simultaneously... i and D j Then the matrix element M ij The value is incremented by 1. When i = j, it records the number of times the running data itself is retrieved.

[0024] Calculate joint probability In the formula, M 次数 For running data D i and D j The number of co-occurrences, where N is the total number of retrievals;

[0025] The second data module constructs a second data set based on the first data set, and the second data module is connected to the first data module.

[0026] The buffer data module constructs a first buffer dataset and a second buffer dataset based on a preset buffer threshold, a first data set, and a second data set. The buffer data module is connected to both the second communication module and the second data module. The process of constructing the first buffer dataset and the second buffer dataset is as follows:

[0027] Based on historical data retrieval records, obtain the retrieval ratio of the first and second historical data sets;

[0028] Within the preset threshold of the buffer, according to the retrieval ratio, the corresponding proportion of running data in the first data set and the second data set are retrieved respectively to construct the first buffer dataset and the second buffer dataset.

[0029] The data update module is used to update historical running data and retrieve records. The data update module is connected to both the second communication module and the first data module.

[0030] The database module is used to store the received information, and it is connected to both the buffer data module and the data update module.

[0031] The present invention is further configured such that: the data management unit includes a third communication module, an information display module, and a preset information module, wherein:

[0032] The third communication module is used to realize information interaction between the data management unit, the data acquisition unit, and the data processing unit;

[0033] The information display module is used to display the received information, and the information display module is connected to the third communication module;

[0034] The preset information module is used to upload preset threshold information, and the preset information module is connected to both the third communication module and the information display module.

[0035] A second aspect of the present invention also provides a method for managing network device operation data based on artificial intelligence, comprising the following steps:

[0036] Collect operational data from network devices and preprocess the collected operational data;

[0037] Based on the historical operation data and retrieval records of network devices, a first dataset is constructed;

[0038] Based on the first data set, a second data set is constructed among the collected operational data.

[0039] Based on the preset buffer threshold, the constructed first data set, and the constructed second data set, a first buffer dataset and a second buffer dataset are constructed respectively.

[0040] Update historical operational data and retrieve records to obtain updated first and second buffer datasets, thus completing the management of network device operational data.

[0041] The present invention is further configured such that the preprocessing process is as follows:

[0042] The operational data is assigned a type identifier based on the network device type from which it is collected;

[0043] Then, the collected operational data is assigned a collection time identifier.

[0044] The present invention is further configured such that the process of constructing the first data set is as follows:

[0045] Obtain historical running data retrieved within a historical data period T, along with the corresponding number of retrievals, and calculate the retrieval probability for each set of running data. In the formula, Ci The number of times the i-th type of running data is retrieved within period T;

[0046] Calculate the mutual information between various running data. In the formula, X and Y are the running data D. i and D j The set of possible values ​​for p(x, y) is D. i =x and D j =The joint probability of y, p(x) and p(y) are respectively D i =x and D j =marginal probability of y;

[0047] The historical running data is sorted according to the number of times it is retrieved, and implicit relationships between the sorted historical running data are established based on the mutual information between the running data, thus obtaining the first data set.

[0048] The present invention is further configured such that the calculation process of the joint probability is as follows:

[0049] Create an n×n matrix M to record the number of co-occurrences among different running data;

[0050] Iterate through all data retrieval records. For each retrieval operation, if the running data D is retrieved simultaneously... i and D j Then the matrix element M ij The value is incremented by 1. When i = j, it records the number of times the running data itself is retrieved.

[0051] Calculate joint probability In the formula, M 次数 For running data D i and D j The number of co-occurrences, where N is the total number of retrievals.

[0052] The present invention is further configured such that the process of constructing the second data set among the collected running data is as follows:

[0053] Based on the number of data type retrievals and implicit associations in the first data set, a second data set is constructed among the collected and running data.

[0054] The present invention is further configured such that the process of constructing the first buffer dataset and the second buffer dataset is as follows:

[0055] Based on historical data retrieval records, obtain the retrieval ratio of the first and second historical data sets;

[0056] Within the preset threshold of the buffer, according to the retrieval ratio, the corresponding proportion of running data in the first data set and the second data set are retrieved respectively to construct the first buffer dataset and the second buffer dataset.

[0057] Compared with the prior art, the beneficial effects of the present invention are:

[0058] This invention calculates the retrieval probability of running data and the mutual information between various running data, and establishes implicit associations to form a first data set and a second data set. This breaks the limitation of traditional data management, which only focuses on the data itself and ignores the relationships between data. Based on the buffer preset threshold, the first data set, and the second data set, a first buffer dataset and a second buffer dataset are constructed, and these buffer datasets are updated in real time. Frequently used data and related data can be pre-stored in the buffer area. When running data needs to be retrieved, the data response speed is greatly improved, the data reading delay is reduced, and the frequency and correlation of data usage are fully considered, which improves the overall performance of the system and the efficiency of data retrieval. At the same time, the continuous updating of historical running data and retrieval records enables the buffer dataset to reflect the latest data situation in real time, realizing a dynamic data buffer storage effect. Attached Figure Description

[0059] Figure 1 This is a system diagram of a network device operation data management system based on artificial intelligence according to the present invention.

[0060] Figure 2 This is a system diagram of the data acquisition unit in a network device operation data management system based on artificial intelligence according to the present invention.

[0061] Figure 3 This is a system diagram of the data processing unit in a network device operation data management system based on artificial intelligence according to the present invention.

[0062] Figure 4 This is a system diagram of the data management unit in a network device operation data management system based on artificial intelligence according to the present invention.

[0063] Explanation of icon numbers:

[0064] 100. Data acquisition unit; 110. Data collection module; 120. Data identification module; 130. First communication module; 200. Data processing unit; 210. Second communication module; 220. First data module; 230. Second data module; 240. Buffered data module; 250. Data update module; 260. Database module; 300. Data management unit; 310. Third communication module; 320. Information display module; 330. Preset information module. Detailed Implementation

[0065] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0066] Example:

[0067] like Figures 1-4 As shown, this embodiment provides a network device operation data management system based on artificial intelligence, including a data acquisition unit 100, a data processing unit 200, and a data management unit 300. The data acquisition unit 100 is used to collect network device operation data and preprocess the collected operation data. The data processing unit 200 is used to construct a first data set, and based on the constructed first data set, construct a second data set among the collected operation data. It also constructs a first buffer dataset and a second buffer dataset based on a buffer preset threshold, the constructed first data set, and the second data set, and updates the first and second buffer datasets. The data processing unit 200 is connected to the data acquisition unit 100. The data management unit 300 is used to display the received information and the preset buffer threshold. The data management unit 300 is connected to both the data acquisition unit 100 and the data processing unit 200.

[0068] In this embodiment, it should be noted that the data acquisition unit 100 uses the operating data of each network device and assigns type and time identifiers to the collected operating data to enable subsequent identification and processing of the operating data. Then, the operating data is uploaded to the data processing unit 200. After receiving the operating data, the data processing unit 200 uses historical operating data information to construct a first data set, and constructs a second data set based on the constructed first data set. Then, based on the buffer preset threshold, the constructed first data set and the second data set, it constructs a first buffer dataset and a second buffer dataset, and updates the first buffer dataset and the second buffer dataset. The first buffer dataset and the second buffer dataset are buffered and stored. When it is necessary to retrieve operating data, the response speed can be effectively improved. At the same time, it can provide buffer storage of corresponding related data, further improving the response speed of retrieving related data. Administrators can use the data management unit 300 to preset the buffer preset threshold in advance, thereby effectively ensuring that the buffered operating data does not exceed the storage threshold, causing buffering delays, and ensuring the stability and reliability of operating data retrieval.

[0069] In this invention, the data acquisition unit 100 includes a data acquisition module 110, a data identification module 120, and a first communication module 130, wherein: the data acquisition module 110 is used to acquire operational data information of network devices; the data identification module 120 is used to assign type and time identifiers to the acquired operational data, and the data identification module 120 is connected to the data acquisition module 110; the first communication module 130 is used to realize information interaction between the data acquisition unit 100, the data processing unit 200, and the data management unit 300.

[0070] In this embodiment, it should be noted that the data acquisition module 110 collects the network device's operating data information and uploads it to the data identification module 120. The data identification module 120 assigns type and time identifiers to the collected operating data and uploads the operating data information with type and time identifiers to the data processing unit 200 and the data management unit 300 through the first communication module 130.

[0071] In this invention, the data processing unit 200 includes a second communication module 210, a first data module 220, a second data module 230, a buffer data module 240, a data update module 250, and a database module 260. The second communication module 210 is used to realize information interaction between the data processing unit 200, the data acquisition unit 100, and the data management unit 300. The first data module 220 is used to construct a first data set, and the first data module 220 is connected to the second communication module 210. The process of constructing the first data set is as follows:

[0072] Obtain historical running data retrieved within a historical data period T, along with the corresponding number of retrievals, and calculate the retrieval probability for each set of running data. In the formula, C i The number of times the i-th type of running data is retrieved within period T;

[0073] Calculate the mutual information between various running data. In the formula, X and Y are the running data D. i and D j The set of possible values ​​for p(x, y) is D. i =x and D j =The joint probability of y, p(x) and p(y) are respectively D i =x and D j =marginal probability of y;

[0074] The historical running data is sorted according to the number of times it is retrieved, and implicit relationships between the sorted historical running data are established based on the mutual information between the running data to obtain the first data set.

[0075] The calculation process for the joint probability is as follows:

[0076] Create an n×n matrix M to record the number of co-occurrences among different running data;

[0077] Iterate through all data retrieval records. For each retrieval operation, if the running data D is retrieved simultaneously... i and D j Then the matrix element M ij The value is incremented by 1. When i = j, it records the number of times the running data itself is retrieved.

[0078] Calculate joint probability In the formula, M 次数 For running data D i and D j The number of co-occurrences, where N is the total number of retrievals;

[0079] The second data module 230 constructs a second data set based on the constructed first data set, and is connected to the first data module 220. The buffer data module 240 constructs a first buffer dataset and a second buffer dataset based on a preset buffer threshold, the constructed first data set, and the second data set. The buffer data module 240 is connected to both the second communication module 210 and the second data module 230. The process of constructing the first buffer dataset and the second buffer dataset is as follows:

[0080] Based on historical data retrieval records, obtain the retrieval ratio of the first and second historical data sets;

[0081] Within the preset threshold of the buffer, according to the retrieval ratio, the corresponding proportion of running data in the first data set and the second data set are retrieved respectively to construct the first buffer dataset and the second buffer dataset.

[0082] The data update module 250 is used to update historical running data and retrieve records. The data update module 250 is connected to both the second communication module 210 and the first data module 220. The database module 260 is used to store the received information. The database module 260 is connected to both the buffer data module 240 and the data update module 250.

[0083] In this embodiment, it should be noted that the second communication module 210 receives the running data information from the data acquisition unit 100 and transmits it to the first data module 220. The first data module 220 calculates the retrieval probability of each running data and the mutual information between each running data based on the historical running data retrieved within the historical data period T and the corresponding retrieval count. It sorts the historical running data according to the retrieval count and establishes implicit associations between the sorted historical running data based on the mutual information between each running data, thus obtaining a first data set. This enables the sorting of data based on retrieval probability and the construction of implicit associations between running data. Then, the second data module 220... Module 230 constructs a second data set of collected data based on the constructed first data set, and transmits the first and second data sets to the buffer data module 240. The buffer data module 240 constructs a first buffer dataset and a second buffer dataset based on the buffer preset threshold, the constructed first data set, and the second data set. This effectively improves the response speed when running data needs to be retrieved. At the same time, it can provide buffer storage of corresponding related data, further improving the response speed of retrieving related data. In addition, the data update module 250 updates historical running data and retrieval records, thereby realizing real-time updates of buffer data to achieve dynamic data buffer storage.

[0084] In this invention, the data management unit 300 includes a third communication module 310, an information display module 320, and a preset information module 330, wherein: the third communication module 310 is used to realize information interaction between the data management unit 300, the data acquisition unit 100, and the data processing unit 200; the information display module 320 is used to display the received information, and the information display module 320 is connected to the third communication module 310; the preset information module 330 is used to upload preset threshold information, and the preset information module 330 is connected to both the third communication module 310 and the information display module 320.

[0085] In this embodiment, it should be noted that the administrator can preset the buffer and preset the threshold through the preset information module 330, upload it to the data processing unit 200 through the third communication module 310, and store it in the database module 260. At the same time, the information display module 320 can display the received information for the administrator to view.

[0086] Furthermore, this embodiment also provides a method for managing network device operation data based on artificial intelligence, including the following steps:

[0087] S1. Collect operational data from network devices and preprocess the collected operational data.

[0088] The preprocessing process is as follows:

[0089] The operational data is assigned a type identifier based on the network device type from which it is collected;

[0090] Then, the collected operational data is assigned a collection time identifier.

[0091] In this embodiment, it should be noted that by collecting the operating data of network devices, raw information reflecting the actual working status of network devices can be obtained. Preprocessing the collected operating data, namely assigning an operating data type identifier and a collection time identifier, makes it easier and more accurate to identify and process this operating data in the future, thereby improving the orderliness and operability of the data.

[0092] S2. Based on the historical operation data and retrieval records of network devices, construct the first data set.

[0093] The process of constructing the first dataset is as follows:

[0094] Obtain historical running data retrieved within a historical data period T, along with the corresponding number of retrievals, and calculate the retrieval probability for each set of running data. In the formula, C i The number of times the i-th type of running data is retrieved within period T;

[0095] Calculate the mutual information between various running data. In the formula, X and Y are the running data D. i and D j The set of possible values ​​for p(x, y) is D. i =x and D j =The joint probability of y, p(x) and p(y) are respectively D i =x and D j =marginal probability of y;

[0096] The historical running data is sorted according to the number of times it is retrieved, and implicit relationships between the sorted historical running data are established based on the mutual information between the running data, thus obtaining the first data set.

[0097] Furthermore, the calculation process for the joint probability is as follows:

[0098] Create an n×n matrix M to record the number of co-occurrences among different running data;

[0099] Iterate through all data retrieval records. For each retrieval operation, if the running data D is retrieved simultaneously... i and D j Then the matrix element M ij The value is incremented by 1. When i = j, it records the number of times the running data itself is retrieved.

[0100] Calculate joint probability In the formula, M 次数 For running data D i and D j The number of co-occurrences, where N is the total number of retrievals.

[0101] In this embodiment, it should be noted that by acquiring historical running data and the corresponding number of times it is retrieved within a historical data period T, and calculating the retrieval probability of each running data, the frequency of use of different running data can be clarified. This helps to understand which data is frequently used and which is relatively infrequently used. Calculating the mutual information between each running data, and establishing and sorting the implicit associations between historical running data based on the mutual information, yields the first data set. This allows for the discovery of potential relationships between running data, providing a more valuable data structure for subsequent data buffer storage and associated data retrieval.

[0102] As an example, suppose a network device operation data scenario involves 5 different types of operation data, denoted as A, B, C, D, and E, and the historical data period T is 10 days. The statistics of the number of times each type of operation data is retrieved within the T period are shown in Table 1.

[0103] Table 1: Statistics of Number of Retrievals

[0104]

[0105]

[0106] The total number of retrievals is 20 + 15 + 10 + 8 + 7 = 60. Calculate the retrieval probability for each running data:

[0107] P A ≈0.333, P B =0.25, P C ≈0.167, P D ≈0.133, P E ≈0.177.

[0108] Create a 5x5 matrix M to record the co-occurrence frequency among different data sets. Assume that after iterating through all data records, the resulting matrix M is as follows:

[0109]

[0110] Calculate the mutual information between A and B if p(A, B) ≈ 0.083.

[0111] The large mutual information indicates a strong correlation between them. The implicit correlation information, combined with the sorted running data, forms the first data set, which contains not only the sorting information of the running data but also the implicit correlation information between them.

[0112] S3. Based on the first data set, construct the second data set among the collected running data.

[0113] The process of constructing the second dataset among the collected operational data is as follows:

[0114] Based on the number of data type retrievals and implicit associations in the first data set, a second data set is constructed among the collected and running data.

[0115] In this embodiment, it should be noted that, based on the first data set, a second data set is constructed by referring to the number of times the data types are retrieved and the implicit relationships therein. This makes the relationships between the collected operational data more comprehensive and detailed. When more information related to the current operational data needs to be obtained, the second data set can provide a more comprehensive reference, which helps to conduct a deeper analysis and understanding of the network device's operation and improves the utilization value of the data.

[0116] S4. Based on the buffer preset threshold, the constructed first data set, and the constructed second data set, construct the first buffer dataset and the second buffer dataset respectively.

[0117] The process of constructing the first buffer dataset and the second buffer dataset is as follows:

[0118] Based on historical data retrieval records, obtain the retrieval ratio of the first and second historical data sets;

[0119] Within the preset threshold of the buffer, according to the retrieval ratio, the corresponding proportion of running data in the first data set and the second data set are retrieved respectively to construct the first buffer dataset and the second buffer dataset.

[0120] In this embodiment, it should be noted that the retrieval ratio of the first data set and the second data set is obtained based on historical data retrieval records. Within the buffer preset threshold, running data is retrieved from the two data sets according to this ratio to construct a buffer dataset. This allows frequently retrieved and related data to be pre-stored in the buffer dataset while meeting storage resource limitations (buffer preset threshold). When running data needs to be retrieved, it can be quickly retrieved directly from the buffer dataset, greatly improving the response speed of data retrieval, reducing the time delay of data reading, and improving the overall performance of the system. At the same time, it can also provide buffer storage for corresponding related data, facilitating quick retrieval of related data.

[0121] S5. Update historical operation data and retrieve records to obtain updated first and second buffer datasets, thus completing the management of network device operation data.

[0122] In this embodiment, it should be noted that continuously updating historical running data and retrieving records ensures that the first and second buffer datasets always reflect the latest data situation. As network devices operate, new data is constantly generated, and the usage of data is also changing. Through real-time updates, the timeliness and accuracy of buffer data are guaranteed, and dynamic data buffer storage is realized.

[0123] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0124] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A network device operation data management system based on artificial intelligence, characterized in that, It includes a data acquisition unit (100), a data processing unit (200), and a data management unit (300), wherein: The data acquisition unit (100) is used to collect the operating data of the network device and preprocess the collected operating data; The data processing unit (200) is used to construct a first data set, and based on the constructed first data set, construct a second data set between the collected running data, and based on the buffer preset threshold, the constructed first data set and the second data set, construct a first buffer dataset and a second buffer dataset respectively, and update the first buffer dataset and the second buffer dataset. The data processing unit (200) is connected to the data acquisition unit (100). The process of constructing the first data set is as follows: Obtain historical running data retrieved within a historical data period T, along with the corresponding number of retrievals, and calculate the retrieval probability for each set of running data. In the formula, , The number of times the i-th type of running data is retrieved within period T; Calculate the mutual information between various running data. In the formula, X and Y are the running data. and The set of possible values, for and The joint probability, , They are respectively and The marginal probability; The historical running data is sorted according to the number of times it is retrieved, and implicit relationships between the sorted historical running data are established based on the mutual information between the running data to obtain the first data set. The calculation process for the joint probability is as follows: Create matrix This is used to record the number of times different runtime data co-occurs; Iterate through all data retrieval records. For each retrieval operation, if runtime data is retrieved simultaneously... and Then the matrix elements The value is incremented by 1 when At that time, it records the number of times the runtime data itself was retrieved; Calculate joint probability In the formula, For running data and The number of times they co-occur. This represents the total number of retrievals. The process of constructing the second dataset among the collected runtime data is as follows: Based on the number of data type retrievals and implicit associations in the first data set, a second data set is constructed among the collected running data. The process of constructing the first buffer dataset and the second buffer dataset is as follows: Based on historical data retrieval records, obtain the retrieval ratio of the first and second historical data sets; Within the preset threshold of the buffer, according to the retrieval ratio, the corresponding proportion of running data in the first data set and the second data set are retrieved respectively to construct the first buffer dataset and the second buffer dataset. The data management unit (300) is used to display the received information and the preset buffer preset threshold. The data management unit (300) is connected to both the data acquisition unit (100) and the data processing unit (200).

2. The network device operation data management system based on artificial intelligence according to claim 1, characterized in that, The data acquisition unit (100) includes a data acquisition module (110), a data identification module (120), and a first communication module (130), wherein: The data acquisition module (110) is used to collect operational data information of network devices; The data identification module (120) is used to assign type and time identifiers to the collected running data. The data identification module (120) is connected to the data acquisition module (110). The first communication module (130) is used to realize information interaction between the data acquisition unit (100), the data processing unit (200), and the data management unit (300).

3. The network device operation data management system based on artificial intelligence according to claim 1, characterized in that, The data processing unit (200) includes a second communication module (210), a first data module (220), a second data module (230), a buffer data module (240), a data update module (250), and a database module (260), wherein: The second communication module (210) is used to realize information interaction between the data processing unit (200), the data acquisition unit (100), and the data management unit (300); The first data module (220) is used to construct a first data set, and the first data module (220) is connected to the second communication module (210); The second data module (230) constructs a second data set between the collected running data based on the constructed first data set, and the second data module (230) is connected to the first data module (220); The buffer data module (240) constructs a first buffer dataset and a second buffer dataset based on a buffer preset threshold, a first data set, and a second data set. The buffer data module (240) is connected to both the second communication module (210) and the second data module (230). The data update module (250) is used to update historical running data and retrieve records. The data update module (250) is connected to both the second communication module (210) and the first data module (220). The database module (260) is used to store the received information, and the database module (260) is connected to both the buffer data module (240) and the data update module (250).

4. The network device operation data management system based on artificial intelligence according to claim 1, characterized in that, The data management unit (300) includes a third communication module (310), an information display module (320), and a preset information module (330), wherein: The third communication module (310) is used to realize information interaction between the data management unit (300), the data acquisition unit (100), and the data processing unit (200); The information display module (320) is used to display the received information, and the information display module (320) is connected to the third communication module (310); The preset information module (330) is used to upload preset threshold information. The preset information module (330) is connected to both the third communication module (310) and the information display module (320).

5. A method for managing network device operation data based on artificial intelligence, characterized in that, Includes the following steps: Collect operational data from network devices and preprocess the collected operational data; Based on the historical operation data and retrieval records of network devices, a first dataset is constructed; Based on the first data set, a second data set is constructed among the collected operational data. Based on the preset buffer threshold, the constructed first data set, and the constructed second data set, a first buffer dataset and a second buffer dataset are constructed respectively. Update historical operational data and retrieve records to obtain updated first and second buffer datasets, thus completing the management of network device operational data; The process of constructing the first data set is as follows: Obtain historical running data retrieved within a historical data period T, along with the corresponding number of retrievals, and calculate the retrieval probability for each set of running data. In the formula, , The number of times the i-th type of running data is retrieved within period T; Calculate the mutual information between various running data. In the formula, X and Y are the running data. and The set of possible values, for and The joint probability, , They are respectively and The marginal probability; The historical running data is sorted according to the number of times it is retrieved, and implicit relationships between the sorted historical running data are established based on the mutual information between the running data to obtain the first data set. The calculation process for the joint probability is as follows: Create matrix This is used to record the number of times different runtime data co-occurs; Iterate through all data retrieval records. For each retrieval operation, if runtime data is retrieved simultaneously... and Then the matrix elements The value is incremented by 1 when At that time, it records the number of times the runtime data itself was retrieved; Calculate joint probability In the formula, For running data and The number of times they co-occur. This represents the total number of retrievals. The process of constructing the second data set among the collected operational data is as follows: Based on the number of data type retrievals and implicit associations in the first data set, a second data set is constructed among the collected running data. The process of constructing the first buffer dataset and the second buffer dataset is as follows: Based on historical data retrieval records, obtain the retrieval ratio of the first and second historical data sets; Within the preset threshold of the buffer, according to the retrieval ratio, the corresponding proportion of running data in the first data set and the second data set are retrieved respectively to construct the first buffer dataset and the second buffer dataset.

6. The method for managing network device operation data based on artificial intelligence according to claim 5, characterized in that, The preprocessing process is as follows: The operational data is assigned a type identifier based on the network device type from which it is collected; Then, the collected operational data is assigned a collection time identifier.

Images