An adjustment method, device, apparatus, medium and program product

Abstract

This application discloses an adjustment method, apparatus, device, medium, and program product; comprising: acquiring initial analysis results; wherein the initial analysis results include results of analyzing the path connectivity status between at least some network nodes in a target network; acquiring user adjustment instruction information; and adjusting the initial analysis results based on the adjustment instruction information using an adjustment model to obtain a target adjustment result. This solution enables flexible and efficient adjustment of the initial analysis results, thereby meeting the diverse data analysis needs of users.

Description

Technical Field

[0001] This application relates to the field of network technology, and in particular to an adjustment method, apparatus, device, medium, and program product. Background Technology

[0002] In practical applications, path analysis results are obtained by analyzing the path status between nodes in a network. However, the path analysis results described above cannot meet the diverse data analysis needs of users. Summary of the Invention

[0003] Based on the above technical problems, embodiments of this application provide an adjustment method, apparatus, device, medium, and program product.

[0004] The technical solution provided in this application is as follows:

[0005] This application first provides an adjustment method, the method comprising:

[0006] Obtain initial analysis results; wherein, the initial analysis results include the results of analyzing the path connectivity status between at least some network nodes in the target network;

[0007] Obtain the user's adjustment instructions;

[0008] The target adjustment result is obtained by adjusting the initial analysis result based on the adjustment instruction information by adjusting the model.

[0009] This application embodiment also provides an adjustment device, the adjustment device comprising:

[0010] The acquisition module is used to acquire initial analysis results and user adjustment instructions; wherein, the initial analysis results include the results of analyzing the path connectivity status between at least some network nodes in the target network;

[0011] The adjustment module is used to adjust the initial analysis results based on the adjustment instruction information using an adjustment model to obtain the target adjustment result.

[0012] This application also provides an electronic device, which includes a processor and a memory; wherein the memory stores a computer program; when the computer program is executed by the processor, it can implement the adjustment method as described above.

[0013] This application also provides a computer-readable storage medium storing a computer program; when the computer program is executed by a processor of an electronic device, it can implement the adjustment method as described above.

[0014] This application also provides a computer program product, which includes a computer program; when the computer program is executed by a processor of an electronic device, it can implement the adjustment method as described above.

[0015] The adjustment method provided in this application, after obtaining initial analysis results from analyzing the path connectivity status between at least some network nodes in the target network and acquiring user adjustment instruction information, adjusts the initial analysis results based on the adjustment instruction information using an adjustment model to obtain the target adjustment result. Thus, through the above operations, automated and intelligent adjustment of the initial analysis results is achieved, improving the matching degree between the target adjustment result and the user's adjustment instruction information. Furthermore, since the adjustment instruction information is added as a factor in the adjustment process of the initial analysis results, targeted adjustments to the initial analysis results can be achieved, ensuring that the target adjustment result obtained through the above adjustments matches the user's adjustment needs. Simultaneously, when the adjustment instruction information changes, the above operations enable diversified and flexible adjustments to the initial analysis results. On the other hand, leveraging the efficiency of the adjustment model in data processing, flexible and efficient adjustments to the initial analysis results can be achieved, thereby meeting the diverse data analysis needs of users. Attached Figure Description

[0016] Figure 1 A flowchart illustrating the adjustment method provided in an embodiment of this application;

[0017] Figure 2 This is a schematic diagram of the structure of the adjustment device provided in the embodiments of this application;

[0018] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0020] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0021] In practical applications, path analysis results are obtained by analyzing the path status between nodes in a network. However, the path analysis results described above cannot meet the diverse data analysis needs of users.

[0022] Based on the above technical problems, embodiments of this application provide an adjustment method, apparatus, device, medium, and program product.

[0023] Figure 1A flowchart illustrating the adjustment method provided in the embodiments of this application is shown below. Figure 1 As shown, the method may include the following steps:

[0024] Step 101: Obtain initial analysis results.

[0025] The initial analysis results include the results of analyzing the path connectivity between at least some network nodes in the target network.

[0026] In one implementation, the target network may include a public network or a private network; exemplarily, the target network may include multiple network nodes; exemplarily, network nodes may include virtual machine devices and / or physical machine devices.

[0027] In one implementation, at least some of the network nodes may be pre-selected or determined, or may be adjusted according to actual network connectivity testing needs.

[0028] In one implementation, network nodes may include cloud hosts for providing Elastic Compute Service (ECS) functionality and related cloud product equipment.

[0029] In one implementation, the path connectivity status may include at least one of the following: connectivity between any two nodes in at least a subset of network nodes, one-way connectivity, or two-way connectivity.

[0030] In one implementation, the initial analysis results may include data such as whether at least some network nodes in the target network are in a connected state, the path transmission delay between at least some network nodes, and the packet loss rate during data transmission between at least some network nodes.

[0031] In one implementation, the initial analysis results can be determined in the following way:

[0032] The tracking results are obtained by tracking the transmission process of traffic data in the target network, and the tracking results are analyzed and integrated to obtain the initial analysis results.

[0033] Step 102: Obtain the user's adjustment instructions.

[0034] In one implementation, the adjustment instruction information may include user-inputted instruction information that characterizes the adjustment of the initial analysis results; for example, the adjustment instruction information may be input in the form of text, voice, and gestures, and this application embodiment does not limit this.

[0035] Step 103: Adjust the initial analysis results based on the adjustment instruction information by adjusting the model to obtain the target adjustment result.

[0036] In one implementation, the tuning model can be a pre-trained neural network model specifically designed to tune and optimize the initial analysis results.

[0037] In one implementation, the target adjustment result may differ from the initial analysis result in at least one dimension, such as data structure, presentation format, and data volume.

[0038] In one implementation, the target adjustment result can be obtained in any of the following ways:

[0039] Based on the adjustment instruction information, the control adjustment model filters the path connectivity status associated with the target node in the initial analysis results to obtain the first result, and determines the first result as the target adjustment result; wherein, the target node can be at least some network nodes among some network nodes.

[0040] Based on the data structure information contained in the adjustment instruction, the control adjustment model adjusts the data structure of the initial analysis results to obtain the target adjustment result.

[0041] As can be seen from the above, the adjustment method provided in this application, after obtaining the initial analysis results obtained by analyzing the path connectivity status between at least some network nodes in the target network and obtaining the user's adjustment instruction information, adjusts the initial analysis results based on the adjustment instruction information using an adjustment model to obtain the target adjustment result. Thus, through the above operations, automated and intelligent adjustment of the initial analysis results is achieved, improving the matching degree between the target adjustment result and the user's adjustment instruction information. Furthermore, since the adjustment instruction information is added as a factor in the adjustment process of the initial analysis results, targeted adjustment of the initial analysis results can be achieved, ensuring that the target adjustment result obtained through the above adjustments matches the user's adjustment needs. Simultaneously, when the adjustment instruction information changes, the above operations enable diversified and flexible adjustments to the initial analysis results. On the other hand, thanks to the efficiency of the adjustment model in the data processing dimension, efficient adjustment of the initial analysis results can be achieved, thereby meeting the diverse data analysis needs of users.

[0042] Based on the foregoing embodiments, the adjustment method provided in this application includes at least a feature extraction module and a sentiment analysis module in the adjustment model.

[0043] In one implementation, the feature extraction module may include a convolutional neural network (CNN); for example, a CNN may include convolutional layers, pooling layers, and fully connected layers.

[0044] In one implementation, the emotion recognition module may include a neural network model with emotion tendency analysis capabilities.

[0045] Accordingly, by adjusting the initial analysis results based on the adjustment instruction information, the target adjustment result can be obtained through the following steps:

[0046] Step A1: Perform pattern recognition on the initial analysis results using the feature extraction module to obtain the pattern recognition results.

[0047] In one implementation, the pattern recognition result may include the result of identification or prediction of the network transmission pattern or network structure pattern characterized by the initial analysis result.

[0048] In one implementation, the pattern recognition result can be obtained in the following way:

[0049] The initial analysis results are preprocessed to obtain preprocessed results. The feature extraction module extracts features of the path connectivity states between different network nodes contained in the preprocessed results to obtain a second result. The second results are then integrated to obtain the pattern recognition result. The data format corresponding to the preprocessed results can be the data format required by the feature extraction module.

[0050] Step A2: Analyze the adjustment instruction information through the sentiment analysis module to obtain the sentiment analysis results.

[0051] In one implementation, the sentiment analysis results may include the sentiment tendency represented by the adjusted indication information; for example, the sentiment tendency may include categories such as positive tendency, negative tendency, and neutrality.

[0052] In one implementation, the sentiment analysis results can be obtained in the following way:

[0053] The sentiment analysis module performs semantic recognition and contextual feature extraction on the adjustment instruction information to obtain a third result. The contextual and semantic features in the third result are then integrated to obtain the sentiment analysis result.

[0054] Step A3: Based on the sentiment analysis results and pattern recognition results, determine the adjustment strategy.

[0055] In one implementation, the adjustment strategy may include at least one of the methods, steps, and procedures for adjusting the initial analysis results.

[0056] In one implementation, the adjustment strategy can be determined in the following way:

[0057] Based on the sentiment analysis results, determine the user's satisfaction level with the initial analysis results and the user's expected results. Then, determine the adjustment strategy based on the degree of satisfaction and the degree of difference between the expected results and the pattern recognition results. For example, if the satisfaction level is unsatisfactory and the expected results do not match the pattern recognition results, the adjustment strategy can be determined as follows: adjust the initial analysis results in target steps until the degree of matching between the adjusted initial analysis results and the expected results is greater than or equal to the matching threshold.

[0058] Step A4: Adjust the initial analysis results based on the adjustment strategy to obtain the target adjustment result.

[0059] In one implementation, the target adjustment result can be obtained in the following way:

[0060] Based on the set of steps and processes included in the adjustment strategy, the initial analysis results are gradually adjusted to obtain the target adjustment result.

[0061] As can be seen from the above, in the adjustment method provided in this application embodiment, the initial analysis result is pattern recognized by the feature extraction module in the adjustment model to obtain the pattern recognition result. In this way, by leveraging the superiority of the feature extraction module in the feature extraction dimension, the accuracy of the pattern recognition result can be improved, as well as the efficiency of obtaining the pattern recognition result. Furthermore, the adjustment instruction information is analyzed by the sentiment analysis module to obtain the sentiment analysis result, which enables the targeted extraction of the sentiment category in the adjustment instruction information. Based on this, an adjustment strategy is determined based on the sentiment analysis result and the pattern recognition result, which can improve the accuracy of the adjustment strategy. Thus, the initial analysis result is adjusted based on the adjustment strategy to obtain the target adjustment result, which not only enables targeted and automated adjustment of the initial analysis result, but also improves the accuracy of the target adjustment result.

[0062] Based on the foregoing embodiments, in the adjustment method provided in this application embodiment, before adjusting the initial analysis results based on the adjustment instruction information using the adjustment model, the following steps may also be performed:

[0063] Step B1: Obtain state sample data.

[0064] The state sample data includes a first set and a second set; the first set includes a set of results for analyzing the path connectivity status between at least some network nodes in at least one network; the second set includes a set of network patterns corresponding to the results in the first set.

[0065] In one implementation, at least one network may include a target network; at least one network may include a local area network (LAN), a metropolitan area network (MAN), or a wide area network (WAN).

[0066] In one implementation, the first set can be predetermined, and correspondingly, the network patterns in the second set can be the label data corresponding to the first set.

[0067] Step B2: Obtain indicator sample data.

[0068] The indicator sample data includes a set of interrelated adjustment indicator information sentiment labels.

[0069] In one implementation, the adjustment instruction information in the indicator sample data may include a set of data for adjusting the results in the first set.

[0070] In one implementation, the sentiment label in the indicator sample data may include the sentiment tendency corresponding to the adjustment indication information in the indicator sample data, and may also include the target pattern of the analysis results of the path connectivity status between at least some network nodes in at least one network, as expected by the user.

[0071] In one implementation, the sentiment labels in the indicator sample data can be preset.

[0072] Step B3: Based on the first set and the second set of state sample data, train the feature extraction module for the initial state to obtain the feature extraction module.

[0073] In one implementation, the feature extraction module can be obtained in the following way:

[0074] The k-th result in the first set is input into the feature extraction module of the initial state, so that the feature extraction module of the initial state can perform pattern recognition on the k-th result in the first set to obtain the k-th recognition result. Then, based on the matching degree between the k-th recognition result and the k-th network pattern in the second set that corresponds to the k-th result in the first set, the parameters of the feature extraction module of the initial state are adjusted until the matching degree is greater than or equal to the matching degree threshold. At this time, the feature extraction module with adjusted parameters can be determined as the feature extraction module; where k is an integer greater than or equal to 1.

[0075] For example, the matching degree can be determined by calculating the kth identification result and the kth network pattern in the second set corresponding to the kth result in the first set using a first loss function.

[0076] Step B4: Based on the adjustment indication information and sentiment labels in the indication sample data, adjust the parameters of the initial state of the sentiment analysis module to obtain the sentiment analysis module.

[0077] In one implementation, the sentiment analysis module can be obtained in the following way:

[0078] The m-th adjustment instruction information from the indicator sample data is input into the initial state sentiment analysis module for analysis, resulting in the m-th sentiment analysis result. The m-th sentiment analysis result and the m-th sentiment label are calculated using a second loss function to determine the degree of difference between the m-th sentiment analysis result and the m-th sentiment label. Based on this degree of difference, the parameters of the initial state sentiment analysis module are adjusted until the degree of difference is less than or equal to the difference threshold. At this point, the initial state sentiment analysis module with adjusted parameters can be defined as the sentiment analysis module. Here, m is an integer greater than or equal to 1, and the m-th sentiment label can be the sentiment label in the indicator sample data corresponding to the m-th adjustment instruction information.

[0079] It should be noted that in the above training process, the state sample data can be divided into a first training dataset and a first verification dataset. The feature extraction module of the initial state is trained based on the first training dataset, and the pattern recognition accuracy of the feature extraction module is verified based on the first verification dataset.

[0080] Accordingly, the indicator sample data can be divided into a second training dataset and a second verification dataset. The sentiment analysis module in the initial state can be trained based on the second training dataset, and the sentiment analysis accuracy of the sentiment analysis module can be verified based on the second verification dataset.

[0081] Step B5: Integrate the feature extraction module and the sentiment analysis module to obtain the adjusted model.

[0082] In one implementation, the feature extraction and sentiment analysis modules can be connected in parallel to obtain an adjusted model.

[0083] In one implementation, the adjustment model may further include a software module for generating adjustment strategies.

[0084] As can be seen from the above, in the adjustment method provided in this application embodiment, after obtaining state sample data, the feature extraction module of the initial state is trained based on the first set and the second set in the state sample data to obtain the feature extraction module. The first set includes a set of results for analyzing the path connectivity state between at least some network nodes in at least one network, and the second set includes a set of network patterns corresponding to the results in the first set. In this way, the supervised training of the feature extraction module of the initial state is realized through the above method, which can not only improve the generalization of the feature extraction module, but also improve the accuracy of the pattern recognition function of the feature extraction module. Furthermore, after obtaining indicator sample data, the parameters of the sentiment analysis module of the initial state are adjusted based on the adjustment indicator information and sentiment tags in the indicator sample data to obtain the sentiment analysis module. The indicator sample data includes a set of interrelated adjustment indicator information and sentiment tags. In this way, targeted training of the sentiment analysis module of the initial state is realized, which can improve the targeting of the sentiment analysis module in performing sentiment analysis on the adjustment indicator information.

[0085] Based on the foregoing embodiments, the adjustment method provided in this application can obtain the initial analysis results in the following ways:

[0086] The transmission status of traffic data in the target network is tracked using a counting Bloom filter; the initial analysis results are determined based on the transmission status.

[0087] Among them, traffic data carries valid data.

[0088] In one implementation, traffic data may include a set of data packets that are actually transmitted by the target network and carry valid data.

[0089] In one implementation, the valid data may correspond to at least one business processing scenario.

[0090] In one implementation, the transmission status may include whether traffic data is transmitted to at least one network node in the target network, and may also include the process of traffic data being transmitted between at least some network nodes.

[0091] In one implementation, at least some network nodes in the target network can be associated with counting Bloom filters. In this way, the counting Bloom filters associated with each network node can detect the state of each network node sending and receiving traffic data, and determine the above-mentioned transmission state based on the state of traffic data being sent and received between each network node.

[0092] In one implementation, the initial analysis results can be determined in the following way:

[0093] The transmission status can include the node identifiers and time information of network nodes. Thus, the transmission process of network data in the target network can be sorted out based on the time information and node identifiers to obtain the fourth result. At the same time, the network topology of the target network is obtained, and the path connectivity status between network nodes in the network topology is identified based on the fourth result to obtain the initial analysis result. The network topology can include a network topology graph, which can be a graphical representation of the target network and can intuitively show the connection relationship between various network nodes in the target network.

[0094] In practical applications, path analysis in cyberspace is an important network management task, used to determine whether the path between a data packet and its destination is connected.

[0095] In related technologies, path analysis methods in cyberspace are mainly based on Internet Control Message Protocol (ICMP) echo requests and connectivity checks. This method typically involves the source device sending ICMP echo request packets, using the target cloud host's Internet Protocol Address (IP) as the destination address, and setting an appropriate time-to-live (TTL). The target cloud host listens for and receives these requests and responds to them. By recording the sending and receiving times, the round-trip time can be calculated, thus determining the path connectivity of the data packets from the source host to the target host.

[0096] Related technologies also provide a path analysis method based on the connectivity of inspection items. This method detects connectivity by judging specific inspection items, including peer-to-peer connections and Internet Content Provider (ICP) registration.

[0097] However, the above method requires sending real ICMP packets to determine the path connectivity status between device nodes, which will generate additional load and interference on the network, and may also expose the real network topology, posing a security risk. Furthermore, the above method cannot fully cover all device nodes and connections in the network, resulting in blind spots in the path connectivity status obtained by the above method. At the same time, the above method cannot detect the path connectivity status of dynamic network structures.

[0098] The adjustment method provided in this application uses a counting Bloom filter to track the transmission status of traffic data in the target network and determines the initial analysis result based on the transmission status, while the traffic data carries valid data. Thus, through the above operations, the network load caused by the need to send specific data packets, including ICMP packets, to analyze path connectivity is reduced; furthermore, by tracking the actual transmission status of traffic data in the target network, diversified, flexible, and dynamic automated tracking of the target network can be achieved, thereby improving the accuracy of the initial analysis result; simultaneously, the risk of leaking the target network topology or leading to network attacks is reduced, thereby improving the security of the process of determining the initial analysis result.

[0099] Based on the foregoing embodiments, the adjustment method provided in this application, which tracks the transmission status of traffic data in the target network using a counting Bloom filter, can be achieved through the following steps:

[0100] Step C1: If a network node in the target network receives the p-th data in the traffic data, it performs a segmentation operation on the p-th data to obtain a set of data blocks.

[0101] Where p is an integer greater than or equal to 1.

[0102] Accordingly, if a network node in the target network does not receive the p-th data in the traffic data, then the p-th data can be left unsegmented.

[0103] In one implementation, the p-th data may include at least one data packet.

[0104] In one implementation, the p-th data can be segmented as follows:

[0105] Construct a counting Bloom filter and perform data splitting operation on the p-th data through the counting Bloom filter to obtain a set of data blocks, denoted as n_gram; for example, the set of data blocks n_gram may include n data block elements; n can be an integer greater than 1.

[0106] Step C2: Encode the data blocks in the data block set to obtain the first encoded set.

[0107] In one implementation, the amount of encoded data in the first encoding set may be less than the amount of data in the corresponding data block in the data block set.

[0108] In one implementation, the first encoding set can be obtained in the following way:

[0109] The data blocks in the data block set are hashed using the hash functions in the hash function set of the count Bloom filter to obtain the first encoding set; correspondingly, the first encoding set may include the set of hash values ​​calculated by the hash functions in the hash function set.

[0110] Step C3: Process the first encoded set using a count Bloom filter to obtain the data block statistics.

[0111] In one implementation, the data block statistics can characterize the number of times a network node receives the p-th data.

[0112] In one implementation, the data block statistics can be obtained in any of the following ways:

[0113] Based on the correspondence between the hash functions in the hash function set and the storage locations of the elements in the count Bloom filter, the encoding results in the first encoding set are stored in the count Bloom filter, and the encoding results stored in the count Bloom filter are statistically analyzed to obtain the data block statistics.

[0114] The statistical results of the data block are obtained by summing the elements corresponding to the p-th data in the count Bloom filter.

[0115] Step C4: Based on the statistical results of data blocks associated with different network nodes, determine the transmission status of traffic data between different network nodes.

[0116] In one implementation, each network node in the target network can be equipped with a counting Bloom filter. In this way, the transmission status of the data blocks corresponding to the data contained in the traffic data can be counted at each network node through the counting Bloom filter corresponding to each network node.

[0117] As can be seen from the above, the adjustment method provided in this application, if a network node in the target network receives the p-th data in the traffic data, performs a segmentation operation on the p-th data to obtain a data block set, encodes the data blocks in the data block set to obtain a first encoding set, then processes the first encoding set through a counting Bloom filter to obtain data block statistics results, and then determines the transmission status of traffic data between different network nodes based on the data block statistics results associated with different network nodes. Thus, in the above process, by leveraging the advantages of the counting Bloom filter in terms of space efficiency, query efficiency, and adaptability to dynamic data, the efficiency of obtaining data block statistics results can be improved, the accuracy of data block statistics results can be improved, and consequently, the accuracy of transmission status and its determination efficiency can be improved.

[0118] Based on the foregoing embodiments, in the adjustment method provided in this application, after obtaining the data block statistical results, the following operations can also be performed:

[0119] If the statistical result of the data block is greater than the preset threshold, it is determined that the p-th data corresponding to the statistical result of the data block is in a threat state.

[0120] Correspondingly, if the statistical results of the data block are less than the preset threshold, it can be determined that the p-th data block is not under threat.

[0121] In one implementation, if the statistical result of the data block is greater than a preset threshold, it can be characterized that the p-th data is in a state of frequent transmission. For example, it can be characterized as a state of threat that may be in the event of private data leakage or frequent attacks.

[0122] As can be seen from the above, the adjustment method provided in this application determines that the p-th data corresponding to the data block statistics is in a threat state if the statistical result of the data block is greater than a preset threshold. Thus, through the above operations, threat detection of traffic data transmitted by at least some network nodes in the target network is achieved.

[0123] Based on the foregoing embodiments, in the adjustment method provided in this application, the transmission state includes the connectivity state between different network nodes in the target network.

[0124] Accordingly, after determining the initial analysis results based on the transmission status, the following operations can also be performed:

[0125] Obtain the path connectivity status with other network nodes in the target network, excluding at least some network nodes; incrementally update the initial analysis results based at least on the path connectivity status with other network nodes.

[0126] In one implementation, the path connectivity status of other network nodes can be obtained after the initial analysis results are determined, or it can be obtained before or during the process of obtaining the initial analysis results.

[0127] In one implementation, the path connectivity status corresponding to other network nodes may include the connectivity status between network nodes in other network nodes, and / or the connectivity status between network nodes in other network nodes and at least some network nodes.

[0128] In one implementation, an incremental update module based on a differential algorithm can be constructed to incrementally update the initial analysis results, thereby reducing update overhead.

[0129] Specifically, the data structure of the incremental update module can be pre-designed, and a persistent data structure can be used to record the initial analysis results and the connectivity status with other network nodes. The above data structure can effectively store the initial analysis results and achieve efficient incremental update operations.

[0130] Secondly, a differential algorithm can be implemented to compare the cached initial analysis results with the connectivity states of other network nodes, and to determine the difference data between the initial analysis results and the aforementioned connectivity states.

[0131] Then, the incremental information corresponding to the above difference data can be incrementally updated into the initial analysis results through the difference algorithm, thereby completing the incremental update of the initial analysis results.

[0132] As can be seen from the above, the adjustment method provided in this application, after obtaining the path connectivity status corresponding to other network nodes in the target network (excluding at least some network nodes), incrementally updates the initial analysis results based at least on the path connectivity status corresponding to other network nodes. Through the above operations, incremental updates to the initial analysis results are achieved, thereby reducing the storage space consumption caused by re-storing the path connectivity status corresponding to other network nodes.

[0133] Based on the foregoing embodiments, the adjustment method provided in this application can also perform the following operations:

[0134] Compress the initial analysis results and / or target adjustment results to obtain compressed data; store the compressed data.

[0135] In one implementation, the amount of compressed data can be less than the amount of data in the initial analysis results, and also less than the amount of data in the target adjustment results.

[0136] In one implementation, compression of the initial analysis results and / or target adjustment results can be achieved in the following way:

[0137] Determine a compression algorithm and compress the initial analysis results and / or target adjustment results using the compression algorithm to obtain compressed data; the compression algorithm may include a string table compression algorithm (Lempel-Ziv-Welch, LZW).

[0138] As can be seen from the above, the adjustment method provided in this application, by compressing the initial analysis results and / or the target adjustment results to obtain and store compressed data, can reduce the storage space consumption caused by storing a large number of initial analysis results and / or target adjustment results.

[0139] Based on the foregoing embodiments, the adjustment method provided in this application, which compresses the initial analysis results and / or the target adjustment results, can be achieved through the following steps:

[0140] Step D1: Determine the second encoding set.

[0141] In one implementation, the second encoding set may include a set of elements such as characters or numbers; for example, the amount of data of character or number elements in the second encoding set may be less than the amount of data in the initial analysis results and / or the target adjustment results.

[0142] In one implementation, the second encoding set can be predetermined.

[0143] Step D2: Construct a dynamic dictionary based on traffic data and the second encoding set.

[0144] In one implementation, the dynamic dictionary may include elements from traffic data and elements from a second encoding set, and there may be a one-to-one correspondence between the elements in the traffic data and the elements in the second encoding set in the dynamic dictionary.

[0145] In one implementation, a dynamic dictionary can be constructed in the following way:

[0146] Initialize the dynamic dictionary by adding elements from the set of data blocks represented by the n-gram corresponding to the p-th data in the traffic data. If the (p+1)-th data in the traffic data is received, determine the set of data blocks corresponding to the (p+1)-th data. If the set of data blocks corresponding to the (p+1)-th data does not exist in the dynamic dictionary, add the set of data blocks corresponding to the (p+1)-th data to the dynamic dictionary. If the set of data blocks corresponding to the (p+1)-th data exists in the dynamic dictionary, expand the (p+2)-th data based on the characters in the (p+2)-th data and check if the expanded (p+1)-th data exists in the dynamic dictionary. If it does not exist, add the expanded (p+1)-th data to the dynamic dictionary. Repeat the above operations recursively until all data blocks corresponding to all data in the traffic data have been processed. At this point, the dynamic dictionary is obtained. Here, p is an integer greater than or equal to 1.

[0147] Step D3: Based on the initial analysis results of dynamic dictionary compression and / or the target adjustment results.

[0148] In one implementation, compressing the initial analysis results and / or the target adjustment results can be achieved in the following way:

[0149] Obtain the q-th data from the initial analysis results and / or target adjustment results. Search for the q-th data in the dynamic dictionary. If the q-th data exists in the dynamic dictionary, obtain the (q+1)-th data. Expand the q-th data based on the characters in the (q+1)-th data to obtain the (q+i)-th expanded result. Then check if the (q+i)-th expanded result exists in the dynamic dictionary. Repeat the above operation until the (q+i)-th expanded result does not exist in the dynamic dictionary. At this point, the encoding element in the dynamic dictionary corresponding to the (q+i)-th expanded result can be determined as the encoding result corresponding to the (q+i)-th expanded result. Repeat this process until all data in the initial analysis results and / or target adjustment results are encoded. Here, q is an integer greater than or equal to 1, i is an integer greater than or equal to 1 and less than or equal to 1, and 1 is an integer greater than or equal to 2.

[0150] As can be seen from the above, the adjustment method provided in this application, after determining the second encoding set, constructs a dynamic dictionary based on traffic data and the second encoding set, and compresses the initial analysis results and / or target adjustment results based on the dynamic dictionary. Thus, through the flexibility, dynamism, scalability, and high efficiency of data operations provided by the dynamic dictionary, comprehensive, stable, and efficient compression processing of the initial analysis results and / or target adjustment results can be achieved; furthermore, through the above compression processing, the probability of data explosion caused by storing the initial analysis results and / or target adjustment results can be reduced.

[0151] It should be noted that, in this embodiment, data sources including log data and performance index data of the target network can be integrated to analyze the path connectivity status between network nodes in the target network. The initial analysis results can also be adjusted and optimized using the aforementioned data sources. Meanwhile, the technical solution provided in this embodiment can also be applied to path analysis across multi-cloud environments to provide data support for managing and optimizing network paths across cloud environments. On the other hand, when performing sentiment analysis, natural language processing technology can be used to analyze feedback data such as network logs and user adjustment support information to uncover more path optimization suggestions and user needs.

[0152] Based on the foregoing embodiments, this application also provides an adjustment device. Figure 2 This is a schematic diagram of the structure of the adjustment device provided in the embodiments of this application, such as... Figure 2 As shown, the adjustment device 2 may include:

[0153] The acquisition module 201 is used to acquire initial analysis results and user adjustment instructions; wherein, the initial analysis results include the results of analyzing the path connectivity status between at least some network nodes in the target network;

[0154] The adjustment module 202 is used to adjust the initial analysis results based on the adjustment instruction information through the adjustment model to obtain the target adjustment result.

[0155] In some embodiments, the adjustment model includes at least a feature extraction module and a sentiment analysis module; the adjustment device may further include a processing module, which is used to perform pattern recognition on the initial analysis results through the feature extraction module to obtain pattern recognition results; analyze the adjustment instruction information through the sentiment analysis module to obtain sentiment analysis results; determine an adjustment strategy based on the sentiment analysis results and the pattern recognition results; and adjust the initial analysis results based on the adjustment strategy to obtain the target adjustment result.

[0156] In some embodiments, the acquisition module 201 is used to acquire state sample data; wherein, the state sample data includes a first set and a second set; the first set includes a set of results of analyzing the path connectivity status between at least some network nodes in at least one network; the second set includes a set of network patterns corresponding to the results in the first set;

[0157] The acquisition module 201 is also used to acquire indicator sample data; wherein, the indicator sample data includes a set of interrelated adjustment indicator information and sentiment labels;

[0158] The processing module is used to train the feature extraction module of the initial state based on the first set and the second set of state sample data to obtain the feature extraction module; to adjust the parameters of the sentiment analysis module of the initial state based on the adjustment indication information and sentiment labels in the indication sample data to obtain the sentiment analysis module; and to integrate the feature extraction module and the sentiment analysis module to obtain the adjustment model.

[0159] In some embodiments, the processing module is configured to track the transmission status of traffic data in a target network using a counting Bloom filter; wherein the traffic data carries valid data.

[0160] The processing module is also used to determine the initial analysis results based on the transmission status.

[0161] In some embodiments, the processing module is configured to perform a segmentation operation on the p-th data in the traffic data if a network node in the target network receives the p-th data, to obtain a set of data blocks; where p is an integer greater than or equal to 1.

[0162] The processing module is also used to encode the data blocks in the data block set to obtain a first encoding set; process the first encoding set through a counting Bloom filter to obtain data block statistics; and determine the transmission status of traffic data between different network nodes based on the data block statistics associated with different network nodes.

[0163] In some embodiments, the processing module is configured to determine that the p-th data corresponding to the data block statistical result is in a threat state if the statistical result of the data block is greater than a preset threshold.

[0164] In some embodiments, the transmission status includes the connectivity status between different network nodes in the target network; the acquisition module 201 is used to acquire the path connectivity status corresponding to other network nodes in the target network, excluding at least some of the network nodes.

[0165] The processing module is used to incrementally update the initial analysis results, at least based on the path connectivity status with other network nodes.

[0166] In some embodiments, the processing module is used to compress the initial analysis results and / or target adjustment results to obtain compressed data; and to store the compressed data.

[0167] In some embodiments, the processing module is configured to determine a second encoding set; construct a dynamic dictionary based on traffic data and the second encoding set; and compress the initial analysis results and / or target adjustment results based on the dynamic dictionary.

[0168] Based on the foregoing embodiments, this application also provides an electronic device. Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application, such as... Figure 3 As shown, the electronic device 3 includes a processor 301 and a memory 302; wherein, the memory 302 stores a computer program; when the computer program is executed by the processor 301, it can implement the adjustment method provided in any of the preceding embodiments.

[0169] Based on the foregoing embodiments, this application also provides a computer-readable storage medium storing a computer program; when the computer program is executed by the processor of an electronic device, it can implement the adjustment method provided in any of the preceding embodiments.

[0170] Based on the foregoing embodiments, this application also provides a computer program product, which includes a computer program; when the computer program is executed by the processor of an electronic device, it can implement the adjustment method provided in any of the preceding embodiments.

[0171] The description of the various embodiments above tends to emphasize the differences between the various embodiments. The similarities or similarities between them can be referred to, and for the sake of brevity, they will not be repeated here.

[0172] The methods disclosed in the various method embodiments provided in this application can be arbitrarily combined to obtain new method embodiments without conflict.

[0173] The features disclosed in the various product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0174] The features disclosed in the various method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0175] It should be noted that the aforementioned computer-readable storage media can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM), etc.; or it can be various electronic devices that include one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc.

[0176] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0177] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0178] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware nodes. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0179] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0180] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0181] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0182] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. An adjustment method, characterized in that, The method includes: The transmission status of traffic data in the target network is tracked using a counting Bloom filter; initial analysis results are determined based on the transmission status; wherein, the initial analysis results include the results of analyzing the path connectivity status between at least some network nodes in the target network; Obtain the user's adjustment instructions; The target adjustment result is obtained by adjusting the initial analysis result based on the adjustment instruction information by adjusting the model; The step of tracking the transmission status of traffic data in the target network using a count Bloom filter includes: if a network node in the target network receives the p-th data in the traffic data, performing a segmentation operation on the p-th data to obtain a data block set; where p is an integer greater than or equal to 1; encoding the data blocks in the data block set to obtain a first encoding set; processing the first encoding set through the count Bloom filter to obtain data block statistics; and determining the transmission status of the traffic data between different network nodes based on the data block statistics associated with different network nodes.

2. The method according to claim 1, characterized in that, The adjustment model includes at least a feature extraction module and a sentiment analysis module; The step of adjusting the initial analysis results based on the adjustment instruction information by adjusting the model to obtain the target adjustment result includes: The initial analysis results are processed by the feature extraction module to obtain the pattern recognition results. The sentiment analysis module analyzes the adjustment instruction information to obtain sentiment analysis results. Based on the sentiment analysis results and the pattern recognition results, an adjustment strategy is determined; The initial analysis results are adjusted based on the adjustment strategy to obtain the target adjustment result.

3. The method according to claim 2, characterized in that, Before adjusting the initial analysis results based on the adjustment instruction information by adjusting the model, the method further includes: Acquire state sample data; wherein the state sample data includes a first set and a second set; the first set includes a set of results for analyzing the path connectivity status between at least some network nodes in at least one network; the second set includes a set of network patterns corresponding to the results in the first set; Obtain indicator sample data; wherein, the indicator sample data includes a set of interrelated adjustment indicator information and sentiment tags; Based on the first set and the second set in the state sample data, the feature extraction module of the initial state is trained to obtain the feature extraction module; Based on the adjustment indication information in the indicated sample data and the sentiment label, the parameters of the initial state of the sentiment analysis module are adjusted to obtain the sentiment analysis module; The feature extraction module and the sentiment analysis module are integrated to obtain the adjustment model.

4. The method according to claim 1, characterized in that, The traffic data contains valid data.

5. The method according to claim 1, characterized in that, After obtaining the statistical results of the data blocks, the method further includes: If the statistical result of the data block is greater than a preset threshold, it is determined that the p-th data corresponding to the statistical result of the data block is in a threat state.

6. The method according to claim 1, characterized in that, The transmission status includes the connectivity status between different network nodes in the target network; after determining the initial analysis result based on the transmission status, the method further includes: Obtain the path connectivity status of the target network nodes, excluding at least some of the network nodes. The initial analysis results are updated incrementally, based at least on the path connectivity status corresponding to the other network nodes.

7. The method according to any one of claims 1 to 6, characterized in that, The method further includes: Compress the initial analysis results and / or target adjustment results to obtain compressed data; Store the compressed data.

8. The method according to claim 7, characterized in that, The compression of the initial analysis results and / or target adjustment results includes: Determine the second encoding set; A dynamic dictionary is constructed based on the traffic data and the second encoding set; The initial analysis results and / or target adjustment results are compressed based on the dynamic dictionary.

9. An adjustment device, characterized in that, The adjustment device includes: The acquisition module is used to track the transmission status of traffic data in the target network through a counting Bloom filter; determine initial analysis results based on the transmission status; and acquire user adjustment instruction information; wherein the initial analysis results include the results of analyzing the path connectivity status between at least some network nodes in the target network; The adjustment module is used to adjust the initial analysis results based on the adjustment instruction information using an adjustment model to obtain the target adjustment result; The acquisition module is further configured to, if a network node in the target network receives the p-th data in the traffic data, perform a segmentation operation on the p-th data to obtain a data block set; wherein p is an integer greater than or equal to 1; encode the data blocks in the data block set to obtain a first encoding set; process the first encoding set through the counting Bloom filter to obtain data block statistics; and determine the transmission status of the traffic data between different network nodes based on the data block statistics associated with different network nodes.

10. An electronic device, characterized in that, The electronic device includes a processor and a memory; wherein the memory stores a computer program; when the computer program is executed by the processor, it is capable of implementing the adjustment method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The storage medium stores a computer program; when the computer program is executed by the processor of the electronic device, it is able to implement the adjustment method as described in any one of claims 1 to 8.

12. A computer program product, characterized in that, The program product includes a computer program; when the computer program is executed by a processor of an electronic device, it is capable of implementing the adjustment method as described in any one of claims 1 to 8.

Images