Data processing method and apparatus, and storage medium

Abstract

This application provides a data processing method, apparatus, and storage medium, relating to the field of data processing technology. The method includes: acquiring a data processing request, the data processing request including at least one first data point, the at least one data point being generated within a preset collection period; acquiring associated data information, the associated data information including at least one historical data point and the correlation between every two historical data points; and determining the target probability of each first data point becoming an alarm root cause based on the data processing request and the associated data information; wherein the first data point and the historical data point are alarm data. This improves the efficiency of processing alarm data.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a data processing method, apparatus and storage medium. Background Technology

[0002] During the operation of a wireless communication system, numerous alarms may be generated. These alarms can affect the normal operation of the wireless communication system; therefore, it is necessary to clear them promptly.

[0003] In practical applications, each alarm needs to be manually checked and processed sequentially. However, some alarms may be caused by other alarms, and these alarms themselves do not indicate operational failures. Checking each alarm sequentially results in low efficiency in alarm processing. Summary of the Invention

[0004] This application provides a data processing method, apparatus, and storage medium, which solves the technical problem of low efficiency in processing alarms.

[0005] In a first aspect, embodiments of this application provide a data processing method, the method comprising:

[0006] A data processing request is obtained, the data processing request including at least one first data, the at least one data being generated within a preset collection period;

[0007] Obtain related data information, which includes at least one historical data point and the relationship between every two historical data points;

[0008] Based on the data processing request and the associated data information, determine the target probability of each first data point becoming the root cause of the alarm;

[0009] The first data and the historical data are alarm data.

[0010] In this embodiment, the target probability of each first data point in the data processing request becoming the root cause of the alarm is determined based on the associated data information. This eliminates the need for manual inspection and processing of each alarm sequentially, thus improving the efficiency of alarm processing.

[0011] In some embodiments, determining the target probability that each first data point becomes an alarm root cause based on the data processing request and the associated data information includes:

[0012] In the associated data information, at least one first data information corresponding to each first data is determined;

[0013] Based on at least one piece of first data information corresponding to each first data, determine the first probability that each first data becomes the root cause of the alarm;

[0014] In the associated data information, second data information corresponding to the at least one first data is determined. The second data information includes historical data that matches the at least one first data, historical data that is associated with the historical data that matches the at least one first data, and the association relationship between every two historical data.

[0015] Based on the second data information, determine the second probability that each first data point becomes the root cause of the alarm;

[0016] For any given first data point, the target probability of the first data point becoming the root cause of the alarm is determined based on the first probability corresponding to the first data point and the second probability corresponding to the first data point.

[0017] In this embodiment, when determining the target probability that a first data point becomes the root cause of an alarm, the target probability that each first data point becomes the root cause of an alarm is determined based on a first probability obtained from at least one first data information corresponding to each first data point, and a second probability obtained from second data information. In this way, the target probability can be determined using probabilities obtained in different ways, improving the accuracy of determining the target probability.

[0018] In some embodiments, determining at least one first data piece of information corresponding to each first data piece in the associated data information includes:

[0019] Obtain a first quantity and a second quantity, wherein the first quantity is the total number of data included in the first data information, and the second quantity is the total number of the at least one first data information;

[0020] For any given first data, a second data is determined from the associated data information, wherein the second data is historical data that matches the first data;

[0021] Based on the associated data information, the first quantity, the second quantity, and the second data, at least one first data information corresponding to the first data is determined.

[0022] In this embodiment of the application, at least one first data information corresponding to each first data is determined in the association set, so as to determine the first probability that each first data becomes the root cause of the alarm based on the at least one first data information corresponding to each first data, thereby improving the accuracy of determining the first probability.

[0023] In some embodiments, the first quantity is M, and the second quantity is N; determining at least one first data piece of information corresponding to the first data based on the associated data information, the first quantity, the second quantity, and the second data includes:

[0024] Determine the association graph corresponding to the associated data information. The association graph includes data nodes corresponding to each historical data, connection relationships and connection directions between each data node, and the association probability that there is an association relationship between each data node.

[0025] Based on the association graph corresponding to the associated data information, determine the steady-state probability of each historical data point;

[0026] Based on M, N, and the association probability between the data nodes, starting from the data node corresponding to the second data, and taking the reverse direction of the connection direction as the traversal direction, N subgraphs are obtained by traversing the association graph. Each subgraph includes M data nodes, and the starting point of each subgraph is the data node corresponding to the second data.

[0027] Based on the N subgraphs, at least one first data information is determined, wherein the first data information includes the historical data corresponding to each data node in the corresponding subgraph, and the correlation between every two historical data.

[0028] In this embodiment, for any given first data, multiple historical data related to the first data are identified in the association graph, and multiple subgraphs are generated through different paths. This allows for the identification of historical data related to the first data from multiple perspectives, resulting in more comprehensive data for determining the first probability and improving the accuracy of that determination.

[0029] In some embodiments, for any given first data, determining the first probability that the first data is the root cause of an alarm based on at least one piece of first data information corresponding to the first data includes:

[0030] In the at least one first data information, determine the target quantity of the first data;

[0031] Based on the target quantity and the steady-state probability of the first data, determine the alarm root cause parameters of the first data;

[0032] The reciprocal of the alarm root cause parameter is determined as the first probability that the first data becomes the alarm root cause.

[0033] In this embodiment of the application, the first probability of each first data becoming the root cause of the alarm is determined based on at least one first data information corresponding to each first data. Since the first data information includes historical data that is related to the first data from multiple perspectives, the accuracy of determining the first probability is improved.

[0034] In some embodiments, determining the second probability that each piece of first data becomes an alarm root cause based on the second data information includes:

[0035] Based on the at least one first data, target data information is determined from the second data information;

[0036] Based on the target data information, determine the second probability that each first data point becomes the root cause of the alarm.

[0037] In this embodiment of the application, the target data information includes each first data and historical data that is related to each first data, but does not include other irrelevant data, which simplifies the process of determining the second probability.

[0038] In some embodiments, determining the target data information in the second data information based on the at least one first data includes:

[0039] Based on the second data information, a correlation subgraph is determined. The correlation subgraph includes at least one data node corresponding to the second data, the connection relationship and connection direction between each data node, and the correlation probability that there is a correlation relationship between each data node.

[0040] Determine whether there are independent data nodes among the data nodes;

[0041] If so, then at least one intermediate data corresponding to each independent data node is determined in the second data information, and data nodes corresponding to each intermediate data are added to the associated subgraph to obtain the target subgraph, and the target data information is determined to include the historical data corresponding to each data node in the target subgraph, and the association relationship between each pair of historical data.

[0042] If not, then the target data information is determined to include the historical data corresponding to each data node in the associated subgraph, and the association relationship between every two historical data.

[0043] Wherein, the second data is historical data that matches the at least one first data, and the independent data node does not have any associated data nodes in the associated subgraph.

[0044] In this embodiment, the target data information is determined based on whether there are independent data nodes in the associated subgraph, ensuring that all data in the target data information have related data, thereby improving the accuracy of determining the second probability.

[0045] In some embodiments, determining the second probability that each first data point becomes an alarm root cause based on the target data information includes:

[0046] Determine the subgraph corresponding to the target data information, and the data nodes corresponding to each first data in the subgraph corresponding to the target data information;

[0047] For any first data, based on the connection direction between the data node corresponding to the first data and the adjacent data nodes, determine the first associated data node to which the data node points and the second associated data node to which the data node points among the at least one adjacent data node;

[0048] The historical data corresponding to the first associated data node is determined as the first associated data corresponding to the first data, and the historical data corresponding to the second associated data node is determined as the second associated data corresponding to the first data;

[0049] Based on the first data and the first associated data corresponding to the first data, determine the first association probability between the first data and the first associated data;

[0050] Based on the first data and the second associated data corresponding to the first data, determine the second association probability between the first data and the second associated data;

[0051] From the target data information, determine the steady-state probability of the first data;

[0052] Based on the first association probability corresponding to the first data, the second association probability corresponding to the first data, and the steady-state probability of the first data, the second probability of the first data becoming the root cause of the alarm is determined.

[0053] In some embodiments, determining the second probability that the first data becomes the root cause of an alarm based on the first association probability corresponding to the first data, the second association probability corresponding to the first data, and the steady-state probability of the first data includes:

[0054] Based on the first association probability, determine the out-degree centrality of the first data;

[0055] The in-degree centrality of the first data is determined based on the second association probability.

[0056] The ratio of the out-degree centrality to the in-degree centrality is determined as the median ratio.

[0057] Based on the intermediate ratio and the steady-state probability corresponding to the first data, a second probability is determined that the first data is the root cause of the alarm.

[0058] In this embodiment, a second probability is determined based on target data information. This improves the accuracy of determining the second probability.

[0059] In some embodiments, the method further includes:

[0060] If the first data does not have a corresponding first associated data, then the intermediate ratio is determined to be a first preset value;

[0061] If the first data does not have a corresponding second associated data, then the intermediate ratio is determined to be a second preset value, which is greater than 1.

[0062] In this embodiment, the case where neither the first nor the second associated data exists is taken into account. This improves the flexibility in determining the second probability.

[0063] In some embodiments, determining the target probability of the first data becoming the root cause of an alarm based on the first probability corresponding to the first data and the second probability corresponding to the first data includes:

[0064] The sum of the first probability corresponding to the first data and the second probability corresponding to the first data is determined as the target probability that the first data becomes the root cause of the alarm.

[0065] In some embodiments, the method further includes:

[0066] Based on the target probability of each first data point becoming the root cause of the alarm, the first data points are sorted in order of magnitude to obtain the target ranking;

[0067] According to the target sorting, each first data is checked and processed in turn to eliminate the alarm caused by at least one first data.

[0068] In this embodiment, based on the target ranking, the first data with the highest target probability is prioritized for investigation. If the first data with the highest target probability is the root cause of alarms generated by other first data, the other first data may be automatically eliminated as the first data with the highest target probability is investigated and resolved. This eliminates the need for manual investigation of multiple data to determine the root cause of alarms before processing, thus improving the efficiency of alarm data processing.

[0069] In some embodiments, the method further includes:

[0070] Obtain at least one historical data point;

[0071] Based on the generation time and data type of each historical data, the at least one historical data is classified to obtain at least one historical data information. The generation time of each historical data in the historical data information is within a preset collection period, and the data type of each historical data is the same.

[0072] The associated data information is obtained by performing rule-based association processing on the at least one historical data information through a preset algorithm.

[0073] In this embodiment, all possible alarm data are collected and associated data information is generated before data processing. This facilitates the subsequent determination of the target probability that each first data point is the root cause of the alarm, improving the accuracy of target probability determination.

[0074] Secondly, embodiments of this application provide a data processing apparatus, including:

[0075] The first acquisition unit is used to acquire a data processing request, the data processing request including at least one first data, the at least one data being generated within a preset acquisition period;

[0076] The second acquisition unit is used to acquire associated data information, which includes at least one historical data and the association relationship between every two historical data.

[0077] The determining unit is configured to determine the target probability of each first data becoming an alarm root cause based on the data processing request and the associated data information;

[0078] The first data and the historical data are alarm data.

[0079] Thirdly, embodiments of this application provide a data processing apparatus, including a memory, a transceiver, and a processor:

[0080] A memory for storing computer programs; a transceiver for sending and receiving data under the control of the processor; and a processor for reading the computer programs from the memory and performing the following operations:

[0081] A data processing request is obtained, the data processing request including at least one first data, the at least one data being generated within a preset collection period;

[0082] Obtain related data information, which includes at least one historical data point and the relationship between every two historical data points;

[0083] Based on the data processing request and the associated data information, determine the target probability of each first data point becoming the root cause of the alarm;

[0084] The first data and the historical data are alarm data.

[0085] In some embodiments, the processor is configured to determine the target probability of each first data point becoming an alarm root cause based on the data processing request and the associated data information, including:

[0086] In the associated data information, at least one first data information corresponding to each first data is determined;

[0087] Based on at least one piece of first data information corresponding to each first data, determine the first probability that each first data becomes the root cause of the alarm;

[0088] In the associated data information, second data information corresponding to the at least one first data is determined. The second data information includes historical data that matches the at least one first data, historical data that has an association relationship with the historical data that matches the at least one first data, and the association relationship between every two historical data.

[0089] Based on the second data information, determine the second probability that each first data point becomes the root cause of the alarm;

[0090] For any given first data point, the target probability of the first data point becoming the root cause of the alarm is determined based on the first probability corresponding to the first data point and the second probability corresponding to the first data point.

[0091] In some embodiments, the processor is configured to determine, within the associated data information, at least one first data piece of information corresponding to each first data piece, including:

[0092] Obtain a first quantity and a second quantity, wherein the first quantity is the total number of data included in the first data information, and the second quantity is the total number of the at least one first data information;

[0093] For any given first data, a second data is determined from the associated data information, wherein the second data is historical data that matches the first data;

[0094] Based on the associated data information, the first quantity, the second quantity, and the second data, at least one first data information corresponding to the first data is determined.

[0095] In some embodiments, the first quantity is M, and the second quantity is N; the processor is configured to determine at least one first data piece of information corresponding to the first data based on the associated data information, the first quantity, the second quantity, and the second data, including:

[0096] Determine the association graph corresponding to the associated data information. The association graph includes data nodes corresponding to each historical data, connection relationships and connection directions between each data node, and the association probability that there is an association relationship between each data node.

[0097] Based on the association graph corresponding to the associated data information, determine the steady-state probability of each historical data point;

[0098] Based on M, N, and the association probability between the data nodes, starting from the data node corresponding to the second data, and taking the reverse direction of the connection direction as the traversal direction, N subgraphs are obtained by traversing the association graph. Each subgraph includes M data nodes, and the starting point of each subgraph is the data node corresponding to the second data.

[0099] Based on the N subgraphs, at least one first data information is determined, wherein the first data information includes the historical data corresponding to each data node in the corresponding subgraph, and the correlation between every two historical data.

[0100] In some embodiments, the processor is configured to, for any given first data, determine a first probability that the first data is an alarm root cause based on at least one first data piece of information corresponding to the first data, including:

[0101] In the at least one first data information, determine the target quantity of the first data;

[0102] Based on the target quantity and the steady-state probability of the first data, determine the alarm root cause parameters of the first data;

[0103] The reciprocal of the alarm root cause parameter is determined as the first probability that the first data becomes the alarm root cause.

[0104] In some embodiments, the processor is configured to determine a second probability that each piece of first data is an alarm root cause based on the second data information, including:

[0105] Based on the at least one first data, target data information is determined from the second data information;

[0106] Based on the target data information, determine the second probability that each first data point becomes the root cause of the alarm.

[0107] In some embodiments, the processor is configured to determine target data information from the second data information based on the at least one first data, including:

[0108] Based on the second data information, a correlation subgraph is determined. The correlation subgraph includes at least one data node corresponding to the second data, the connection relationship and connection direction between each data node, and the correlation probability that there is a correlation relationship between each data node.

[0109] Determine whether there are independent data nodes among the data nodes;

[0110] If so, then at least one intermediate data corresponding to each independent data node is determined in the second data information, and data nodes corresponding to each intermediate data are added to the associated subgraph to obtain the target subgraph, and the target data information is determined to include the historical data corresponding to each data node in the target subgraph, and the association relationship between each pair of historical data.

[0111] If not, then the target data information is determined to include the historical data corresponding to each data node in the associated subgraph, and the association relationship between every two historical data.

[0112] Wherein, the second data is historical data that matches the at least one first data, and the independent data node does not have any associated data nodes in the associated subgraph.

[0113] In some embodiments, the processor is configured to determine a second probability that each piece of first data becomes an alarm root cause based on the target data information, including:

[0114] Determine the subgraph corresponding to the target data information, and the data nodes corresponding to each first data in the subgraph corresponding to the target data information;

[0115] For any first data, based on the connection direction between the data node corresponding to the first data and the adjacent data nodes, determine the first associated data node to which the data node points and the second associated data node to which the data node points among the at least one adjacent data node;

[0116] The historical data corresponding to the first associated data node is determined as the first associated data corresponding to the first data, and the historical data corresponding to the second associated data node is determined as the second associated data corresponding to the first data;

[0117] Based on the first data and the first associated data corresponding to the first data, determine the first association probability between the first data and the first associated data;

[0118] Based on the first data and the second associated data corresponding to the first data, determine the second association probability between the first data and the second associated data;

[0119] From the target data information, determine the steady-state probability of the first data;

[0120] Based on the first association probability corresponding to the first data, the second association probability corresponding to the first data, and the steady-state probability of the first data, the second probability of the first data becoming the root cause of the alarm is determined.

[0121] In some embodiments, the processor is configured to determine a second probability that the first data becomes the root cause of an alarm based on a first association probability corresponding to the first data, a second association probability corresponding to the first data, and a steady-state probability of the first data, including:

[0122] Based on the first association probability, determine the out-degree centrality of the first data;

[0123] The in-degree centrality of the first data is determined based on the second association probability.

[0124] The ratio of the out-degree centrality to the in-degree centrality is determined as the median ratio.

[0125] Based on the intermediate ratio and the steady-state probability corresponding to the first data, a second probability is determined that the first data is the root cause of the alarm.

[0126] In some embodiments, the processor is further configured to:

[0127] If the first data does not have a corresponding first associated data, then the intermediate ratio is determined to be a first preset value;

[0128] If the first data does not have a corresponding second associated data, then the intermediate ratio is determined to be a second preset value, which is greater than 1.

[0129] In some embodiments, the processor is configured to determine the target probability of the first data becoming an alarm root cause based on a first probability corresponding to the first data and a second probability corresponding to the first data, including:

[0130] The sum of the first probability corresponding to the first data and the second probability corresponding to the first data is determined as the target probability that the first data becomes the root cause of the alarm.

[0131] In some embodiments, the processor is further configured to:

[0132] Based on the target probability of each first data point becoming the root cause of the alarm, the first data points are sorted in order of magnitude to obtain the target ranking;

[0133] According to the target sorting, each first data is checked and processed in turn to eliminate the alarm caused by at least one first data.

[0134] In some embodiments, the processor is further configured to:

[0135] Obtain at least one historical data point;

[0136] Based on the generation time and data type of each historical data, the at least one historical data is classified to obtain at least one historical data information. The generation time of each historical data in the historical data information is within a preset collection period, and the data type of each historical data is the same.

[0137] The associated data information is obtained by performing rule-based association processing on the at least one historical data information through a preset algorithm.

[0138] Fourthly, embodiments of this application provide a non-transitory readable storage medium storing a computer program for causing a processor to execute the method described in any one of the first aspects.

[0139] Fifthly, embodiments of this application provide a data processing device, wherein the communication device stores a computer program, the computer program being used to cause a processor to execute the method described in any one of the first aspects.

[0140] The data processing method, apparatus, and storage medium provided in this application determine the target probability that each first data in a data processing request becomes the root cause of an alarm based on associated data information. This eliminates the need for manual inspection and processing of each alarm sequentially, thus improving alarm processing efficiency.

[0141] It should be understood that the description in the foregoing summary section is not intended to limit the key or essential features of the embodiments of the present invention, nor is it intended to restrict the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0142] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0143] Figure 1A A schematic diagram of an application scenario provided in this application embodiment;

[0144] Figure 1B Illustration of the application scenarios provided in the embodiments of this application Figure 2 ;

[0145] Figure 2 Flowchart 1 of the data processing method provided in the embodiments of this application;

[0146] Figure 3 Flowchart of the data processing method provided in the embodiments of this application Figure 2 ;

[0147] Figure 4 A schematic diagram of the association diagram provided in the embodiments of this application;

[0148] Figure 5 A schematic diagram illustrating the process of determining at least one first data information provided in an embodiment of this application;

[0149] Figure 6 A schematic diagram of the intermediate association diagram corresponding to the second data information provided in the embodiments of this application;

[0150] Figure 7 This is a schematic diagram illustrating the process of determining target data information provided in an embodiment of this application;

[0151] Figure 8 A schematic diagram illustrating the data processing procedure provided in the embodiments of this application;

[0152] Figure 9 A schematic diagram of the structure of the data processing apparatus provided in the embodiments of this application is shown below;

[0153] Figure 10 Schematic diagram of the data processing apparatus provided in the embodiments of this application Figure 2 ;

[0154] Figure 11 Schematic diagram of the data processing apparatus provided in the embodiments of this application Figure 3 . Detailed Implementation

[0155] In the embodiments of this application, the term "multiple" refers to two or more, and other quantifiers are similar.

[0156] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0157] To facilitate a clear description of the technical solutions in the embodiments of this application, some terms and technologies involved in the embodiments of this application will be briefly introduced below:

[0158] 1. Markov Chain

[0159] A Markov chain is a stochastic process with the Markov property (memoryless), whose future state depends only on the current state and is independent of past states. It is an important modeling tool for time-discrete, state-discrete, or continuous stochastic systems and is widely used in statistics, physics, finance, artificial intelligence, and other fields.

[0160] (1) Important properties

[0161] Irreducibility: A Markov chain is said to be irreducible if it is possible to reach any state j from any state i (within a finite number of steps). Otherwise, there may be multiple closed subsets (such as absorbing states).

[0162] Periodicity: If the number of steps to return to a certain state has a greatest common divisor d > 1, then the state is said to be periodic (e.g., d = 2 indicates alternation). If all states are periodic d = 1, then the chain is said to be aperiodic.

[0163] Recurrence: In a recurring state, the probability that state i will be visited an infinite number of times is 1. Transient state: State i may no longer be visited.

[0164] Stationary distribution: If there exists a probability distribution π such that:

[0165] π = πP (discrete time)

[0166] or,

[0167] πQ = 0 (continuous time)

[0168] Then π is called a stationary distribution. Here, P is the transition probability matrix; Q is the transition rate matrix. An irreducible, aperiodic finite-state Markov chain must have a unique stationary distribution.

[0169] (2) Markov chain Monte Carlo (MCMC) path

[0170] MCMC path refers to the sequence of samples generated by random walks through a Markov chain, which eventually converge to the target probability distribution.

[0171] 2. Directed graph

[0172] A directed graph is a graph structure consisting of a set of vertices and a set of directed edges, where each edge points from one vertex to another. Vertices represent entities (such as web pages, people, or tasks), and directed edges represent unidirectional relationships (such as hyperlinks or dependencies).

[0173] (1) In-degree and out-degree

[0174] In-degree: The number of edges pointing to this vertex.

[0175] Out-degree: The number of edges originating from this vertex.

[0176] (2) Path

[0177] A directed path is a series of vertices along an edge.

[0178] Simple path: A path that does not repeat the vertices.

[0179] A loop is a path where the starting point and the ending point are the same.

[0180] (3) Connectivity

[0181] Strongly connected: There is a bidirectional path between any two vertices.

[0182] Weakly connected: An undirected graph is connected if the direction is ignored.

[0183] Directed acyclic graph: A directed graph without cycles (such as task scheduling dependencies).

[0184] 3. Other terms

[0185] In this embodiment of the invention, the term "and / or" describes the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following associated objects have an "or" relationship.

[0186] In the embodiments of this application, the term "at least one" refers to one or more items, and "more than one" refers to two or more items. Other quantifiers are similar. For example, at least one of a, b, or c can be expressed as: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0187] The terms "first," "second," etc., used in the embodiments of this application are for illustrative purposes and to distinguish the objects being described. They do not indicate any order or limit on the number of objects in the embodiments of this application, and cannot constitute any limitation on the embodiments of this application. For example, the use of terms such as "first random access signal" and "second random access signal" is only to distinguish different random access signals, and does not indicate any difference in the size, priority, or importance of these two random access signals.

[0188] Figure 1A This is a schematic diagram illustrating an application scenario provided in an embodiment of this application. For example... Figure 1A As shown, this includes terminal equipment and base stations. The terminal equipment and base stations can communicate within a wireless communication system. During communication between the terminal equipment and base stations, alarms may be generated due to communication failures or anomalies.

[0189] Figure 1B Illustration of the application scenarios provided in the embodiments of this application Figure 2 .like Figure 1B As shown, this includes a first base station and a second base station. The first base station and the second base station can communicate with each other in a wireless communication system. During the communication between the first base station and the second base station, alarms may be generated due to communication failures or anomalies.

[0190] It should be noted that the above application scenarios and the number of devices in each application scenario are merely examples. For instance, the wireless communication scenario can be other scenarios, and this application does not limit the application scenarios or the number of devices in each application scenario.

[0191] The technical solutions provided in this application can be applied to a variety of systems. For example, applicable systems may include LTE systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, long term evolution advanced (LTE-A) systems, universal mobile telecommunication system (UMTS), worldwide interoperability for microwave access (WiMAX) systems, 5G NR systems and their evolved communication systems, and 6G (sixth generation mobile communication technology) systems. These systems may include terminals and network equipment. The systems may also include a core network component, such as evolved packet core (EPC) and 5G core network (5GC).

[0192] The network device involved in the embodiments of this application can be a base station, which may include multiple cells providing services to the UE. Depending on the specific application, the base station may also be called an access point, or a device in the access network that communicates with the wireless UE through one or more sectors on the air interface, or other names. The network device can be used to exchange received air frames with Internet Protocol (IP) packets, acting as a router between the wireless UE and the rest of the access network, where the rest of the access network may include an Internet Protocol (IP) communication network. The network device can also coordinate the attribute management of the air interface. For example, the network device involved in the embodiments of this application may be an evolved Node B (eNB or e-NodeB) in an LTE system, a 5G base station (gNB) in a 5G network architecture (next generation system), or a home evolved node B (HeNB), relay node, femto, pico, network testing equipment, etc., and is not limited in the embodiments of this application. In some network architectures, network devices may include centralized unit (CU) nodes and distributed unit (DU) nodes, which may also be geographically separated.

[0193] The UE involved in the embodiments of this application can be a device that provides voice and / or data connectivity to a user, a handheld device with wireless connectivity, or other processing devices connected to a wireless modem. The UE can also be referred to as a terminal. A wireless terminal can be a USB storage device, other personal computer memory devices, and a dongle. It can also communicate with one or more core networks (CNs) via a radio access network (RAN). A wireless terminal can be a mobile terminal, such as a mobile phone (or "cellular" phone) and a computer with a mobile terminal. For example, it can be a portable, pocket-sized, handheld, computer-embedded, or vehicle-mounted mobile device that exchanges voice and / or data with the radio access network. Examples include personal communication service (PCS) phones, cordless phones, session initiated protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), personal computers, tablets, machine-type communication (MTC) terminals, and other devices. Wireless terminals can also be referred to as systems, subscriber units, subscriber stations, mobile stations, mobile stations, remote stations, access points, remote terminals, access terminals, user terminals, user agents, user devices, and wireless access devices and routers / modems that meet the limitations of this definition, but are not limited in the embodiments of this application.

[0194] This application provides a data processing method, apparatus, and storage medium. The method of this application can determine the target probability that each first data point in a data processing request becomes the root cause of an alarm based on associated data information. This eliminates the need for manual inspection and processing of each alarm sequentially, improving alarm processing efficiency. The method and apparatus are based on the same concept, and since the principles of solving the problem are similar, their implementations can be mutually referenced; repeated details will not be elaborated further.

[0195] Figure 2 A flowchart of a data processing method provided in an embodiment of this application. For example... Figure 2 As shown, the method includes:

[0196] S201, Obtain data processing request.

[0197] The execution entity in this application embodiment can be a data processing device or a data processing apparatus disposed within a data processing device. The data processing apparatus can be implemented through software or a combination of software and hardware. For example, the data processing device can be a server.

[0198] The data processing request includes at least one first data point, and the generation time of at least one data point is within a preset acquisition period.

[0199] The first data is alarm data. Alarm data is data generated during the operation of the wireless communication system.

[0200] The preset collection cycle can be set to a preset duration. The preset duration can be set in advance and stored in the preset storage space of the data processing device.

[0201] The data processing device can collect alarm data generated during the operation of the wireless communication system and determine the data processing request based on the generation time of each alarm data.

[0202] For example, assuming a preset duration of 1 minute, the preset acquisition period can be determined to be at least one first data point acquired every 1 minute. During the operation of the wireless communication system, the data processing device acquires at least one first data point, including alarm data A1, alarm data A2, and alarm data A3. The generation times of alarm data A1, alarm data A2, and alarm data A3 are between 14:20 and 14:21 on April 1, 2025. The data processing device obtains data processing requests including alarm data A1, alarm data A2, and alarm data A3, which are acquired within the preset acquisition period.

[0203] S202. Obtain related data information.

[0204] The associated data information includes at least one historical data point and the relationship between every two historical data points.

[0205] Historical data can be alarm data generated within a historical period.

[0206] Related data information can be represented using a directed graph. In a directed graph, each data node can be historical data, and the edge between any two data nodes can represent the relationship between any two historical data nodes.

[0207] Suppose we have two historical data sets, namely the first historical data set and the second historical data set, and the edge connecting the two sets points from the first historical data set to the second. Then we can determine that the relationship between the two sets of historical data is that the first historical data set is the root cause of the alarm in the second historical data set. In other words, the cause of the second historical data set is the first historical data set.

[0208] Assume the data node corresponding to the first historical data is designated as Data Node 1, and the data node corresponding to the second historical data is designated as Data Node 2. The connection direction between Data Node 1 and Data Node 2 is from Data Node 1 to Data Node 2. The probability of a correlation between the two data nodes can be represented as the probability that Data Node 1 is the root cause of the alarm generated by Data Node 2.

[0209] Before making a data processing request, the data processing device can generate correlated data information based on multiple historical data. The correlated data information includes multiple alarm data that may be generated during the operation of the wireless communication system, as well as the correlation between each pair of alarm data.

[0210] It should be noted that the "association relationships" involved in the associated data information, first data information, second data information, and target data information in this application can all refer to the relevant descriptions of the association relationships mentioned above, and will not be repeated hereafter. Similarly, the "association probabilities" involved in this application can all refer to the relevant descriptions of the association probabilities mentioned above, and will not be repeated hereafter.

[0211] In some embodiments, associated data information can be generated in the following manner: acquiring at least one historical data; classifying the at least one historical data according to the generation time and data type of each historical data to obtain at least one historical data information, wherein the generation time of each historical data in the historical data information is within a preset collection period and the data type of each historical data is the same; and performing rule association processing on the at least one historical data information through a preset algorithm to obtain associated data information.

[0212] The data type of historical data can include operator type, technology type, and base station type. The operator type indicates the operator to which the base station generating the alarm data belongs. The technology type can be 4G or 5G. The base station type indicates the base station that generated the alarm data.

[0213] At least one set of historical data contains data where the operator type, technology type, and base station type are all identical.

[0214] For example, the historical data collected within the preset collection period includes alarm data B1 to alarm data B5. The specific data type of each alarm data can be shown in Table 1:

[0215] Table 1

[0216]

[0217]

[0218] As shown in Table 1, alarm data B1 to B5 can be divided into two historical data information sets: Historical Data Information 1 and Historical Data Information 2. Historical Data Information 1 includes alarm data B1, alarm data B3, and alarm data B5. Historical Data Information 2 includes alarm data B2 and alarm data B4.

[0219] Optionally, if at least one historical data information contains historical data information that is identical to other historical data information, then that historical data information is deleted.

[0220] If the historical data in two sets of historical data are identical, then it can be determined that the two sets of historical data are identical.

[0221] For example, suppose historical data information 1 includes alarm data B1, alarm data B3, and alarm data B5, and historical data information 3 includes alarm data B1, alarm data B3, and alarm data B5. Then it can be determined that historical data information 1 and historical data information 3 are the same, and historical data information 1 is deleted.

[0222] The default algorithm can be an association rule algorithm. For example, the default algorithm can be an association rule algorithm (Apriori algorithm) or a frequent pattern growth (FP-Growth) algorithm.

[0223] Since the preset algorithm is an association rule algorithm, it can be used to perform rule-based association processing on at least one historical data point to obtain associated data information. The associated data information includes data nodes corresponding to multiple historical data points. The association relationships between data nodes retain only first-order association terms, that is, only the association term between one antecedent and one consequent is retained. The association relationships between data nodes can be represented by association probabilities. The higher the association probability, the greater the probability that an association relationship exists between the data nodes.

[0224] For example, in the process of A→B, A is the antecedent and B is the consequent. The association between A and B can be represented by the association probability.

[0225] After generating the associated data information, the data processing device stores the associated data information in the preset storage space of the data processing device.

[0226] S203. Based on the data processing request and related data information, determine the target probability of each first data becoming the root cause of the alarm.

[0227] In some embodiments, the target probability of each first data point becoming the root cause of an alarm can be determined based on a data processing request and associated data information in the following manner: In the associated data information, at least one first data point corresponding to each first data point is determined; based on the at least one first data point corresponding to each first data point, a first probability of each first data point becoming the root cause of an alarm is determined; in the associated data information, second data point corresponding to at least one first data point is determined, the second data point including historical data matching at least one first data point, historical data having an association relationship with the historical data matching at least one first data point, and the association relationship between every two historical data points; based on the second data point, a second probability of each first data point becoming the root cause of an alarm is determined; for any first data point, based on the first probability corresponding to the first data point and the second probability corresponding to the first data point, the target probability of the first data point becoming the root cause of an alarm is determined.

[0228] If the historical data is the same as the first data, or if the historical data and the first data are of the same data type, then it can be determined that the historical data matches the first data.

[0229] For example, as illustrated above, at least one set of first data is determined to include alarm data A1, alarm data A2, and alarm data A3. Regarding alarm data A1, assuming that alarm data 1 is identical to alarm data A1 in the associated data information, then alarm data 1 matches alarm data A1. Therefore, the historical data matching the first data is alarm data 1. Regarding alarm data A2, assuming that alarm data 2 has the same data type as alarm data A2 in the associated data information, then alarm data 2 matches alarm data A2. Therefore, the historical data matching the first data is alarm data 2.

[0230] For example, as illustrated above, at least one first data point is determined to include alarm data A1, alarm data A2, and alarm data A3. The data processing device, within the associated data information, determines at least one first data point corresponding to each first data point. Based on this at least one first data point, the probability that each first data point is the root cause of an alarm is determined, as shown in Table 2.

[0231] Table 2

[0232] First Data Alarm data A1 Alarm data A2 Alarm data A3 First probability 0.8 0.5 0.4

[0233] The data processing device determines at least one second data information corresponding to a first data in the associated data information, and determines the second probability of each first data becoming the root cause of the alarm based on the second data information, as shown in Table 3:

[0234] Table 3

[0235] First Data Alarm data A1 Alarm data A2 Alarm data A3 Second probability 0.6 0.4 0.35

[0236] For any given first data point, the data processing device determines the target probability that the first data point becomes the root cause of the alarm based on the first probability and the second probability corresponding to the first data point. The specific target probabilities of each first data point becoming the root cause of the alarm are shown in Table 4.

[0237] Table 4

[0238] First Data Alarm data A1 Alarm data A2 Alarm data A3 Target probability 1.4 0.9 0.75

[0239] The data processing equipment can determine the root cause of the alarm by identifying the first data point with the highest target probability, as shown in Table 4. When investigating and processing the first data point, alarm data A1 is processed first. Thus, after processing alarm data A1, if the root cause of alarm data A1 and alarm data A3 is alarm data A1, alarm data A1 and alarm data A3 may be automatically eliminated as alarm A1 is investigated and resolved. This eliminates the need for manual investigation of alarm data A1, alarm data A2, and alarm data A3 sequentially to determine the root cause before processing, thus improving the efficiency of alarm data processing.

[0240] The second data information is the associated data information, which includes at least one first data and the minimum data information that is associated with each first data. In this way, the second probability of each first data becoming the root cause of the alarm can be determined only by the set of data that are associated with each first data, thus improving the accuracy of determining the second probability.

[0241] When determining the target probability that a first data point is the root cause of an alarm, the target probability of each first data point becoming the root cause is determined based on a first probability obtained from at least one first data point corresponding to each first data point, and a second probability obtained from the second data point. In this way, the target probability can be determined using probabilities obtained through different methods, thus improving the accuracy of target probability determination.

[0242] The data processing method provided in this application can determine the target probability that each first data in a data processing request becomes the root cause of an alarm based on associated data information. This eliminates the need for manual inspection and processing of each alarm sequentially, thus improving the efficiency of alarm processing.

[0243] Based on any of the above embodiments, the following, in conjunction with Figure 3 The process of determining the target probability (S203) of each first data point becoming the root cause of the alarm is explained.

[0244] Figure 3 Flowchart of the data processing method provided in the embodiments of this application Figure 2 .like Figure 3 As shown, the method includes:

[0245] S301. In the associated data information, determine at least one first data information corresponding to each first data.

[0246] In some embodiments, at least one first data information corresponding to each first data can be determined in the associated data information in the following manner: obtaining a first quantity and a second quantity, wherein the first quantity is the total number of data included in the first data information, and the second quantity is the total number of at least one first data information; for any first data, determining second data in the associated data information, wherein the second data is historical data matching the first data; and determining at least one first data information corresponding to the first data based on the associated data information, the first quantity, the second quantity, and the second data.

[0247] The first and second quantities can be set in advance and stored in the preset storage space of the data processing device.

[0248] The first quantity is M, and the second quantity is N. In some embodiments, at least one first data information corresponding to the first data can be determined based on the associated data information, the first quantity, the second quantity, and the second data in the following manner: Determine the association graph corresponding to the associated data information, the association graph including data nodes corresponding to each historical data, the connection relationships and connection directions between each data node, and the association probability that there is an association relationship between each data node; determine the steady-state probability of each historical data based on the association graph corresponding to the associated data information; based on M, N, and the association probability that there is an association relationship between each data node, starting from the data node corresponding to the second data and traversing in the reverse direction of the connection direction, traverse the association graph to obtain N subgraphs, each subgraph including M data nodes, with the starting point of the subgraph being the data node corresponding to the second data; determine at least one first data information based on the N subgraphs, the first data information including the historical data corresponding to each data node in the corresponding subgraph, and the association relationship between every two historical data.

[0249] A relationship graph can be a directed graph. Below, we will combine... Figure 4 The relationship diagram will be explained.

[0250] Figure 4 This is a schematic diagram of the association diagram provided for an embodiment of this application. For example... Figure 4As shown, this includes a correlation diagram 401. Correlation diagram 401 includes the data nodes corresponding to alarm data 1 to alarm data 18, the connection relationships and directions between each data node, and the probability of association between the data nodes. Taking alarm data 1 to alarm data 4 as an example, correlation diagram 401 includes the connection relationship and direction between alarm data 1 and alarm data 2, and the probability of association between alarm data 1 and alarm data 2 is 0.6. The connection relationship and direction between alarm data 1 and alarm data 3, and the probability of association between alarm data 1 and alarm data 3 is 0.3. The connection relationship and direction between alarm data 1 and alarm data 4, and the probability of association between alarm data 1 and alarm data 4 is 0.1.

[0251] The steady-state probability of each historical data can be determined based on the association graph corresponding to the associated data information in the following way: Determine the association matrix corresponding to the association graph. The association matrix is ​​an L*L square matrix, where L is the total number of alarm data included in the association graph, and each element in the association matrix is ​​the association probability that there is an association relationship between each data node; determine the steady-state probability of each historical data based on the association probability.

[0252] In the correlation matrix, the (v)th i ,v j The element represents the alarm data v. i With alarm data v j The probability π of an association relationship exists ij Among them, alarm data v i With alarm data v j The connection direction between them is alarm data v i Pointing to alarm data v j If there is no correlation between two alarm data, the corresponding element in the correlation matrix will be 0.

[0253] Before determining the steady-state probability of each historical data point based on the association probability, the association matrix needs to satisfy the nonperiodicity and irreducibility of the Markov chain.

[0254] The aperiodicity of the association graph can be determined as follows: For any alarm data, find all paths that can return to itself. If such a cycle exists, calculate the length of all cycle paths. If the greatest common divisor among these lengths is 1, then the graph satisfies aperiodicity; otherwise, it does not. If no cycle path from any alarm item node to itself exists, then the graph satisfies aperiodicity.

[0255] If an alarm item does not satisfy the non-periodicity requirement, then directly add a loop from the alarm item to itself and assign a decimal close to 0 as the transition probability.

[0256] The irreducibility of an association graph can be determined as follows: traverse each alarm item, using it as a starting point and moving along the transition path. If it can reach any other alarm item, then the graph is irreducible; otherwise, it is not. If the graph is not irreducible, add an edge between the alarm item and unreachable alarm items, assigning a transition probability close to 0. In fact, if there is enough historical alarm data, both aperiodicity and irreducibility are generally satisfied.

[0257] The steady-state probability of historical data can be determined using the following equation:

[0258]

[0259] The constraints are:

[0260]

[0261] Among them, P j For the j-th historical data v j steady-state probability; P i For the i-th historical data v i The steady-state probability. See above for explanations of the other parameters.

[0262] Below, in conjunction with Figure 5 The process of determining at least one first data piece of information is described.

[0263] Figure 5 This is a schematic diagram illustrating the process of determining at least one first data piece of information provided in an embodiment of this application. For example... Figure 5 As shown, multiple alarm data are included, namely alarm data A1 to alarm data A9. Alarm data A1 to alarm data A9 are historical data, and they are in the association graph. Assume the data processing request includes alarm data B1 to alarm data B3. For alarm data B1, assuming that alarm data A4 in the association graph is the same as alarm data B1, then the historical data matching alarm data B1 in the association graph is alarm data A4. Therefore, the data processing device determines the second data as alarm data A4. Assume the first quantity M is 4 and the second quantity N is 4. Starting from the data node corresponding to alarm data A4, and using the reverse direction of the connection direction as the traversal direction, the first data node in the reverse direction of the data node of alarm data A4 is the data node corresponding to alarm data A9.

[0264] The data node corresponding to alarm data A9 can be traversed in reverse to two other data nodes, namely the data nodes corresponding to alarm data A5 and alarm data A6. Since the correlation probability between alarm data A6 and alarm data A9 is greater than that between alarm data A5 and alarm data A9, the data processing device traverses to the data node corresponding to alarm data A6. The data node corresponding to alarm data A6 can also be traversed in reverse to two other data nodes, namely the data nodes corresponding to alarm data A2 and alarm data A3. Since the correlation probability between alarm data A3 and alarm data A6 is greater than that between alarm data A2 and alarm data A6, the data processing device traverses to the data node corresponding to alarm data A3.

[0265] When traversing to alarm data A3, the data processing device determines that the traversed data nodes include alarm data A4, alarm data A9, alarm data A6, and the data node corresponding to alarm data A3. Since four data nodes have been traversed, the data processing device determines that the traversal is complete. Subgraph 1 is generated, which includes four data nodes: alarm data A4, alarm data A9, alarm data A6, and the data node corresponding to alarm data A3. The data processing device determines that the first data information includes alarm data A4, alarm data A9, alarm data A6, alarm data A3, and the correlation between every two alarm data points.

[0266] Since it is necessary to traverse to obtain 4 subgraphs, the data processing device continues to traverse to obtain the subgraphs using the method described above. Figure 2 ,son Figure 2 It includes four data nodes: alarm data A4, alarm data A9, alarm data A6, and the data node corresponding to alarm data A2. The data processing device determines that the first data information includes alarm data A4, alarm data A9, alarm data A6, alarm data A2, and the correlation between every two alarm data.

[0267] The data processing device continues to traverse and obtain sub-data using the above method. Figure 3 ,son Figure 3 It includes four data nodes, namely the data nodes corresponding to alarm data A4, alarm data A9, alarm data A5, and alarm data A6. The data processing device determines that the first data information includes alarm data A4, alarm data A9, alarm data A5, alarm data A6, and the correlation between every two alarm data.

[0268] The data processing device continues to traverse and obtain sub-data using the above method. Figure 4 ,son Figure 4It includes four data nodes: alarm data A4, alarm data A9, alarm data A5, and the data node corresponding to alarm data A3. The data processing device determines that the first data information includes alarm data A4, alarm data A9, alarm data A5, alarm data A3, and the correlation between every two alarm data.

[0269] In summary, the historical data included in at least one first data piece of information corresponding to alarm data A1 can be specifically shown in Table 5:

[0270] Table 5

[0271] First Data Information Historical data Data Information 1 Alarm data A4, Alarm data A9, Alarm data A6, Alarm data A3 Data Information 2 Alarm data A4, Alarm data A9, Alarm data A6, Alarm data A2 Data Information 3 Alarm data A4, Alarm data A9, Alarm data A5, Alarm data A6 Data Information 4 Alarm data A4, Alarm data A9, Alarm data A5, Alarm data A3

[0272] Optionally, if the number of subgraphs obtained through traversal is less than N, then at least one first data information can be generated directly based on the obtained subgraphs.

[0273] For example, assuming N is 4, the data processing device can only obtain 3 sub-graphs according to the above method, and then at least one first data information is generated based on the 3 sub-graphs.

[0274] S302. Based on at least one first data information corresponding to each first data, determine the first probability that each first data becomes the root cause of the alarm.

[0275] In some embodiments, for any first data, the first probability of each first data becoming an alarm root cause can be determined by the following method based on at least one first data information corresponding to each first data: determining the target number of the first data in at least one first data information; determining the alarm root cause parameter of the first data based on the target number and the steady-state probability of the first data; and determining the reciprocal of the alarm root cause parameter as the first probability of the first data becoming an alarm root cause.

[0276] The target number is the total number of first data items that exist in at least one first data item corresponding to each first data item.

[0277] Since the first data and the second data match, the steady-state probability of the second data corresponding to the first data in the association graph can be determined as the steady-state probability of the first data.

[0278] The product of the target quantity and the steady-state probability of the first data can be used to determine the alarm root cause parameter of the first data.

[0279] For example, the data processing device obtains at least one first data piece of information corresponding to each first data piece of information according to the above method, totaling 16 first data pieces of information. The data processing device determines the target number of first data pieces of information to be 6. The data processing device determines that the second data piece corresponding to alarm data B1 is alarm data A4, and therefore determines the steady-state probability of alarm data B1 to be the steady-state probability of alarm data A4, which is 0.2. The data processing device determines the alarm root cause parameter of data B1 to be 0.2 * 8 = 1.6, and determines the first probability of the first data piece becoming the alarm root cause to be 1 / 1.6 = 0.625. According to the above method, the specific first probability of each first data piece becoming the alarm root cause can be determined by the data processing device as shown in Table 6:

[0280] Table 6

[0281] First Data Alarm data B1 Alarm data B2 Alarm data B3 First probability 0.625 0.5 0.4

[0282] S303. In the associated data information, determine at least one second data information corresponding to the first data.

[0283] The second data information includes historical data that matches at least one first data, historical data that is associated with at least one first data, and the association between any two historical data.

[0284] In the association graph corresponding to the associated data information, at least one second data information corresponding to the first data can be identified.

[0285] For example, based on the above example, at least one first data is determined to be alarm data B1 to alarm data B3. The specific second data corresponding to each first data in the association diagram of the linked data information can be shown in Table 7:

[0286] Table 7

[0287] First Data Second data Alarm data B1 Alarm data A4 Alarm data B2 Alarm data A6 Alarm data B3 Alarm Data A5

[0288] According to Table 7, the data processing device determines the second data information in the association diagram. The intermediate association diagram corresponding to the second data information can be specifically as follows: Figure 6 As shown.

[0289] Figure 6 This is a schematic diagram of the intermediate association diagram corresponding to the second data information provided in the embodiments of this application. For example... Figure 6As shown, the intermediate association diagram 601 corresponding to the second data information includes the data nodes corresponding to each second data (data nodes corresponding to alarm data A4, alarm data A5 and alarm data A6) shown in Table 7, the data nodes of historical data that are associated with each data node (data nodes corresponding to alarm data A2, alarm data A3, alarm data A7, alarm data A8 and alarm data A9), the connection relationship and connection direction between each data node, and the association probability that there is an association relationship between each pair of data nodes.

[0290] S304. Based on the second data information, determine the second probability that each first data point is the root cause of the alarm.

[0291] In some embodiments, the second probability of each first data point becoming an alarm root cause can be determined based on the second data information in the following manner: determining target data information in the second data information based on at least one first data point; and determining the second probability of each first data point becoming an alarm root cause based on the target data information.

[0292] In some embodiments, target data information can be determined in second data information based on at least one first data in the following manner: Based on the second data information, a correlation subgraph is determined, the correlation subgraph including at least one data node corresponding to the second data, the connection relationship and connection direction between each data node, and the correlation probability that there is a correlation relationship between each data node; it is determined whether there is an independent data node among each data node; if so, at least one intermediate data corresponding to each independent data node is determined in the second data information, and data nodes corresponding to each intermediate data are added to the correlation subgraph to obtain the target subgraph, and the target data information is determined to include historical data corresponding to each data node in the target subgraph, and the correlation relationship between every two historical data; if not, the target data information is determined to include historical data corresponding to each data node in the correlation subgraph, and the correlation relationship between every two historical data; wherein, the second data is historical data matching at least one first data, and the independent data node does not have a data node with a correlation relationship in the correlation subgraph.

[0293] The associated subgraph can be determined in the intermediate association graph corresponding to the second data information.

[0294] Below, in conjunction with Figure 7 The process of determining the target data information is explained.

[0295] Figure 7 This is a schematic diagram illustrating the process of determining target data information provided in an embodiment of this application. For example... Figure 7 As shown, it includes a related subgraph 701 and a target subgraph 702. The data processing device is... Figure 6In the intermediate association graph corresponding to the second data information shown, association subgraph 701 is determined. Association subgraph 701 includes data nodes corresponding to alarm data A4, alarm data A5, and alarm data A6, the connection relationships and directions between each data node, and the association probability between each data node. The data processing device determines that among the data nodes, there exists an independent data node corresponding to alarm data A4. In the second data information, the data processing device determines that the intermediate data corresponding to alarm data A1 is alarm data A3. Alarm data A3 has association relationships with alarm data A4, alarm data A5, and alarm data A6. The data processing device adds the data node corresponding to alarm data A3 to association subgraph 701 to obtain target subgraph 702. The data processing device determines that the target data information includes the historical data corresponding to each data node in target subgraph 702, and the association relationships between every two historical data. That is, the target data information includes alarm data A3 to alarm data A6, and the association relationships between every two alarm data.

[0296] In some embodiments, the second probability of each first data point becoming the root cause of an alarm can be determined based on the target data information in the following manner: determining the subgraph corresponding to the target data information, and the data nodes corresponding to each first data point in the subgraph corresponding to the target data information; for any first data point, determining, based on the connection direction between the data node corresponding to the first data point and the adjacent data nodes, a first associated data node and a second associated data node pointing to the data node in at least one adjacent data node; determining the historical data corresponding to the first associated data node as the first associated data corresponding to the first data point, and determining the historical data corresponding to the second associated data node as the second associated data corresponding to the first data point; determining the first association probability between the first data point and the first associated data point based on the first data point and the first associated data point corresponding to the first data point; determining the second association probability between the first data point and the second associated data point corresponding to the first data point; determining the steady-state probability of the first data point in the target data information; and determining the second probability of the first data point becoming the root cause of an alarm based on the first association probability, the second association probability, and the steady-state probability of the first data point.

[0297] For example, according to the above Figure 7 As shown, for alarm data B1, the second data corresponding to alarm data B1 is determined to be alarm data A4. Since in target sub-graph 702, there is only a connection between alarm data A4 and alarm data A3, and the connection direction between alarm data A4 and alarm data A3 is from alarm data A4 to alarm data A3, therefore, alarm data B1 corresponding to alarm data A4 only has the first associated data, alarm data A3.

[0298] In some embodiments, the second probability of the first data becoming the root cause of an alarm can be determined by the following method based on the first association probability corresponding to the first data, the second association probability corresponding to the first data, and the steady-state probability of the first data: determining the out-degree centrality corresponding to the first data based on the first association probability; determining the in-degree centrality corresponding to the first data based on the second association probability; determining the ratio of the out-degree centrality to the in-degree centrality as the intermediate ratio; and determining the second probability of the first data becoming the root cause of an alarm based on the intermediate ratio and the steady-state probability corresponding to the first data.

[0299] The out-degree centrality of the first data point can be determined using the following formula 1:

[0300]

[0301] Among them, C out (v k The out-degree centrality of the first data point is denoted as ). Let be the association probability of the k-th outgoing edge corresponding to the first data in the association graph; n is the total number of data in the target data information that are related to the first data.

[0302] For example, as illustrated above, the first associated data of alarm data B1 is determined to be alarm data A3. In the target subgraph 702, since the connection direction corresponding to alarm data A4 points to alarm data A3, and alarm data A4 has only one connection direction, the probability that alarm data A4, corresponding to alarm data B1, has only one corresponding outgoing edge is 0.25, which is the association probability between alarm data A4 and alarm data A3. In the target subgraph 702, the data associated with alarm data A4 is alarm data A3. Therefore, n is 1.

[0303] The in-degree centrality of the first data point can be determined using the following formula 2:

[0304]

[0305] Among them, C in (v k ) represents the in-degree centrality of the first data point; This represents the association probability of the k-th incoming edge corresponding to the first data point. See above for explanations of the other parameters.

[0306] For example, according to Table 7 above, the second data corresponding to alarm data B2 is alarm data A6. Regarding alarm data A6, in the target subgraph 702, since the connection direction of other alarm data pointing to alarm data A6 is alarm data A3, alarm data B2 has a corresponding second associated data, alarm data A3. The association probability of the incoming edge corresponding to the first data is the association probability between alarm data A3 and alarm data A6, which is 0.7. In the target subgraph 702, the data that are associated with alarm data A6 are alarm data A3 and alarm data A5. Therefore, n is 2.

[0307] The intermediate ratio can be determined using the following formula 3:

[0308]

[0309] Where l is the median ratio. See above for explanations of the other parameters.

[0310] Optionally, if the first data does not have a corresponding first associated data, the intermediate ratio is determined to be a first preset value; if the first data does not have a corresponding second associated data, the intermediate ratio is determined to be a second preset value, and the second preset value is greater than 1.

[0311] A first preset value and a second preset value can be set in advance and stored in the preset storage space of the data processing device. The first preset value is 0.

[0312] For example, as illustrated above, it is determined that alarm data B1 only has the first associated data and no corresponding second associated data. Therefore, the median ratio can be determined to be the second preset value of 2.

[0313] In some embodiments, the second probability of the first data becoming the root cause of the alarm can be determined by the following method based on the steady-state probability corresponding to the intermediate ratio and the first data: the product of the steady-state probability of the first data and the intermediate ratio is determined as the intermediate product, and the reciprocal of the intermediate product is determined as the second probability of the first data becoming the root cause of the alarm.

[0314] For example, assuming the data processing device determines the second probability of each first data point becoming an alarm root cause according to the method shown in the above embodiments, it can be specifically as shown in Table 8:

[0315] Table 8

[0316] First Data Alarm data B1 Alarm data B2 Alarm data B3 Second probability 0.5 0.48 0.35

[0317] S305. For any first data, determine the target probability of the first data becoming the root cause of the alarm based on the first probability corresponding to the first data and the second probability corresponding to the first data.

[0318] In some embodiments, the target probability of the first data becoming the root cause of the alarm can be determined by the first probability corresponding to the first data and the second probability corresponding to the first data in the following manner: the sum of the first probability corresponding to the first data and the second probability corresponding to the first data is determined as the target probability of the first data becoming the root cause of the alarm.

[0319] For example, according to Tables 6 and 8 above, the target probability is determined to be 0.625 + 0.5 = 1.125 for alarm data B1. The target probability is determined to be 0.5 + 0.48 = 0.98 for alarm data B2. The target probability is determined to be 0.4 + 0.35 = 0.75 for alarm data B3.

[0320] After determining the target probability of each first data point being the root cause of the alarm, the first data points can be sorted in order of magnitude based on their target probability to obtain the target ranking. Based on the target ranking, each first data point is investigated and processed in turn to eliminate alarms caused by at least one first data point.

[0321] For example, as illustrated above, the target order is determined as alarm data B1, alarm data B2, and alarm data B3. Alarm data B1 is most likely to be the root cause of the alarm. Therefore, the data processing device can first eliminate the faults and anomalies in alarm data B1 based on the target order. After alarm data B1 is resolved, alarm data B2 and alarm data B3 may be automatically cleared.

[0322] Optionally, after obtaining the target sorting, it can be sent to a preset device. This allows users to obtain the target sorting through the preset device and perform manual troubleshooting. This eliminates the need for manual investigation of multiple data points sequentially to determine the root cause of the alarm before processing, thus improving the efficiency of alarm data processing.

[0323] The data processing method provided in this application can determine the target probability of each first data point becoming the root cause of an alarm based on associated data information. If the first data point is the root cause of an alarm generated by other data, the other data may be automatically eliminated as the first data point is investigated and resolved. This eliminates the need for manual investigation of multiple data points sequentially to determine the root cause of the alarm before processing, thus improving the efficiency of alarm data processing.

[0324] Based on any of the above embodiments, the following, in conjunction with Figure 8 The process of data processing will be illustrated with examples.

[0325] Figure 8 This is a schematic diagram illustrating the data processing procedure provided in an embodiment of this application. Figure 8As shown, the system includes a network device 801 and a data processing device 802. The network device 801 can be a base station, and the data processing device 802 can be a server. When the network device 801 communicates in the wireless communication system, it generates multiple service data. If an operational anomaly causes a communication failure, alarm data will be generated. The data processing device 802 can collect the alarm data generated by the network device 801 according to a preset collection period and generate a data processing request. Assuming the preset duration is 1 minute, the preset collection period can be determined as collecting at least one first data point every 1 minute. During the operation of the network device 801, the at least one first data point collected by the data processing device 802 includes alarm data C1, alarm data C2, and alarm data C3. The generation times of alarm data C1, alarm data C2, and alarm data C3 are between 2025 / 04 / 01:15:20 and 2025 / 04 / 01:15:21. The data processing device acquires data processing requests including alarm data C1, alarm data C2, and alarm data C3, which are collected within a preset collection period.

[0326] Data processing device 802 acquires associated data information and determines the associated graph corresponding to the associated data information. Data processing device 802 determines the associated matrix corresponding to the associated graph, where each element in the associated matrix represents the probability of an association relationship between data nodes. Based on the associated probabilities, the steady-state probability of each historical data point in the associated data information is determined. In the associated data information, data processing device 802 identifies historical data matching alarm data C1, alarm data C2, and alarm data C3 as second data, and determines the steady-state probability of the second data corresponding to the first data, as shown in Table 9.

[0327] Table 9

[0328] First Data Second data steady state probability Alarm data C1 Alarm data A1 0.6 Alarm data C2 Alarm data A2 0.75 Alarm data C3 Alarm data A3 0.4

[0329] Data processing device 802 acquires a first quantity M of 4 and a second quantity N of 5. For alarm data C1, based on M, N, and the correlation probability between data nodes, starting from the data node corresponding to alarm data C1 and traversing in the reverse direction of the connection, five subgraphs are obtained in the correlation graph. Each subgraph contains four data nodes. Based on these five subgraphs, at least one first data piece of information is determined. This first data piece of information includes the historical data corresponding to each data node in the corresponding subgraph. The specific details of obtaining at least one first data piece of information corresponding to each first data piece using this method are shown in Figure 10.

[0330] Table 10

[0331] First Data First Data Information Alarm data C1 Data Information 1 to Data Information 5 Alarm data C2 Data Information 6 to Data Information 10 Alarm data C3 Data Information 11 to Data Information 15

[0332] For alarm data C1, data processing device 802 determines the target quantity of alarm data C1 to be 6 from data information 1 to data information 15 shown in Table 10. According to Table 9, the steady-state probability of alarm data C1 is determined to be 0.6. Data processing device 802 determines the alarm root cause parameter to be 0.6 * 6 = 3.6. Data processing device 802 determines the first probability that alarm data C1 becomes the alarm root cause to be the reciprocal of 3.6, 0.278. Based on the above method, data processing device 802 obtains the specific first probabilities of each first data point becoming the alarm root cause, as shown in Table 11.

[0333] Table 11

[0334] First Data Alarm data C1 Alarm data C2 Alarm data C3 First probability 0.278 0.267 0.25

[0335] The data processing device 802 determines the second data information corresponding to alarm data C1 to alarm data C3 from the associated data information. Within the second data information, the data processing device 802 determines an association subgraph. The association subgraph includes the data nodes corresponding to alarm data A1, alarm data A2, and alarm data A3, the connection relationships and directions between each data node, and the association probability of the relationships between the data nodes. If the data processing device 802 determines that there are no independent data nodes in the association subgraph, then the target data information includes the historical data corresponding to each data node in the association subgraph, and the association relationships between every two historical data points.

[0336] For alarm data C1, data processing device 802 determines from the target data information that the first associated data corresponding to alarm data C1 is alarm data A2, and there is no second associated data. Data processing device 802 directly determines the median ratio to be the second preset value of 2. For alarm data C2, data processing device 802 determines from the target data information that the first associated data corresponding to alarm data C1 is alarm data A3, and the second associated data is alarm data A1. Data processing device 802 determines the first association probability to be 0.65 and the second association probability to be 0.7. Based on the first association probability, data processing device 802 determines the out-degree centrality of alarm data C2 using the above formula 1. Based on the second association probability, data processing device 802 determines the in-degree centrality of alarm data C2 using the above formula 2. Data processing device 802 determines the median ratio to be 2.8. For alarm data C3, data processing device 802 determines from the target data information that alarm data C1 has no first associated data, and the second associated data corresponding to alarm data C1 is alarm data A3. Data processing device 802 directly determines the intermediate ratio to be 0.

[0337] For any given first data point, the data processing device 802 determines the intermediate product by multiplying the steady-state probability of the first data point by the intermediate ratio, and then determines the reciprocal of the intermediate product as the second probability that the first data point becomes the root cause of the alarm. The specific second probabilities of each first data point becoming the root cause of the alarm are shown in Table 12.

[0338] Table 12

[0339] First Data Alarm data C1 Alarm data C2 Alarm data C3 Second probability 0.83 0.47 0

[0340] Based on the first probability shown in Table 11 and the second probability shown in Table 12, the data processing device 702 determines the target probability of the first data becoming the root cause of the alarm, as shown in Table 13.

[0341] Table 13

[0342] First Data Alarm data C1 Alarm data C2 Alarm data C3 Target probability 1.108 0.737 0.25

[0343] According to Table 13, the data processing device 802 sorts the first data according to the order of target probability, resulting in the target sorting as alarm data C1, alarm data C2, and alarm data C3. When investigating the alarm data, alarm data C1 is investigated first.

[0344] The data processing procedure provided in this application embodiment can determine the target probability of each first data point becoming the root cause of an alarm based on associated data information. If the first data point is the root cause of an alarm generated by other data, the other data may be automatically eliminated as the first data point is investigated and resolved. This eliminates the need for manual investigation of multiple data points sequentially to determine the root cause of the alarm before processing, thus improving the efficiency of alarm data processing.

[0345] Based on the same technical concept, this application also provides a data processing apparatus. This data processing apparatus can realize the functions of the data processing device in the foregoing embodiments.

[0346] Figure 9 A schematic diagram of the data processing apparatus provided in an embodiment of this application is shown below. Figure 9 As shown, the data processing device 900 may include:

[0347] The first acquisition unit 901 is used to acquire a data processing request. The data processing request includes at least one first data, and the generation time of the at least one data is within a preset acquisition period.

[0348] The second acquisition unit 902 is used to acquire associated data information, which includes at least one historical data and the association relationship between every two historical data.

[0349] The determining unit 903 is used to determine the target probability of each first data becoming the root cause of the alarm based on the data processing request and the associated data information.

[0350] Among them, the first data and historical data are alarm data.

[0351] In some embodiments, the determining unit 903 is specifically used for:

[0352] In the associated data information, at least one first data information corresponding to each first data is determined;

[0353] Based on at least one piece of first data information corresponding to each first data, determine the first probability that each first data becomes the root cause of the alarm;

[0354] In the associated data information, at least one second data information corresponding to the first data is determined. The second data information includes historical data that matches at least one first data, historical data that has an association relationship with the historical data that matches at least one first data, and the association relationship between every two historical data.

[0355] Based on the second data information, determine the second probability that each of the first data points is the root cause of the alarm.

[0356] For any given first data point, determine the target probability that the first data point becomes the root cause of the alarm based on the first probability and the second probability corresponding to the first data point.

[0357] In some embodiments, the determining unit 903 is specifically used for:

[0358] Obtain a first quantity and a second quantity, wherein the first quantity is the total number of data included in the first data information, and the second quantity is the total number of at least one piece of the first data information;

[0359] For any given first data, determine the second data from the associated data information. The second data is historical data that matches the first data.

[0360] Based on the associated data information, the first quantity, the second quantity, and the second data, determine at least one first data information corresponding to the first data.

[0361] In some embodiments, the first quantity is M, the second quantity is N, and the determining unit 903 is specifically used for:

[0362] Determine the association graph corresponding to the associated data information. The association graph includes the data nodes corresponding to each historical data, the connection relationship and connection direction between each data node, and the association probability that there is an association relationship between each data node.

[0363] Based on the correlation diagram corresponding to the correlation data, determine the steady-state probability of each historical data point;

[0364] Based on M, N, and the probability of association between the data nodes, starting from the data node corresponding to the second data, and taking the reverse direction of the connection as the traversal direction, N subgraphs are obtained by traversing the association graph. Each subgraph contains M data nodes, and the starting point of the subgraph is the data node corresponding to the second data.

[0365] Based on N subgraphs, at least one first data information is determined. The first data information includes the historical data corresponding to each data node in the corresponding subgraph, and the relationship between every two historical data.

[0366] In some embodiments, the determining unit 903 is specifically used for:

[0367] In at least one set of first data information, determine the target quantity of the first data;

[0368] Based on the target quantity and the steady-state probability of the first data, determine the alarm root cause parameters of the first data;

[0369] The reciprocal of the alarm root cause parameter is determined as the first probability that the first data becomes the alarm root cause.

[0370] In some embodiments, the determining unit 903 is specifically used for:

[0371] Based on at least one first data, the target data information is determined from the second data information;

[0372] Based on the target data information, determine the second probability that each first data point is the root cause of the alarm.

[0373] In some embodiments, the determining unit 903 is specifically used for:

[0374] Based on the second data information, a correlation subgraph is determined. The correlation subgraph includes at least one data node corresponding to the second data, the connection relationship and connection direction between each data node, and the correlation probability that there is a correlation relationship between each data node.

[0375] Determine whether there are independent data nodes among the data nodes;

[0376] If so, then determine at least one intermediate data corresponding to each independent data node in the second data information, add data nodes corresponding to each intermediate data in the correlation subgraph to obtain the target subgraph, and determine that the target data information includes the historical data corresponding to each data node in the target subgraph, as well as the correlation relationship between each pair of historical data.

[0377] If not, then the target data information includes the historical data corresponding to each data node in the associated subgraph, and the relationship between every two historical data.

[0378] The second data is historical data that matches at least one first data, and independent data nodes do not have any related data nodes in the association subgraph.

[0379] In some embodiments, the determining unit 903 is specifically used for:

[0380] Determine the subgraph corresponding to the target data information, and the data nodes corresponding to the first data in each subgraph corresponding to the target data information;

[0381] For any first data, based on the connection direction between the data node corresponding to the first data and the adjacent data nodes, determine the first associated data node that the data node points to and the second associated data node that points to the data node in at least one adjacent data node.

[0382] The historical data corresponding to the first associated data node is determined as the first associated data corresponding to the first data, and the historical data corresponding to the second associated data node is determined as the second associated data corresponding to the first data;

[0383] Based on the first data and the first associated data corresponding to the first data, determine the first association probability between the first data and the first associated data;

[0384] Based on the first data and the second associated data corresponding to the first data, determine the second association probability between the first data and the second associated data;

[0385] Determine the steady-state probability of the first data point from the target data information;

[0386] Based on the first correlation probability corresponding to the first data, the second correlation probability corresponding to the first data, and the steady-state probability of the first data, the second probability of the first data becoming the root cause of the alarm is determined.

[0387] In some embodiments, the determining unit 903 is specifically used for:

[0388] Determine the out-degree centrality of the first data based on the first association probability;

[0389] The in-degree centrality of the first data is determined based on the second association probability.

[0390] The ratio of out-degree centrality to in-degree centrality is determined as the median ratio.

[0391] Based on the median ratio and the steady-state probability corresponding to the first data, determine the second probability that the first data is the root cause of the alarm.

[0392] In some embodiments, the determining unit 903 is further configured to:

[0393] If the first data does not have a corresponding first associated data, then the intermediate ratio is determined to be the first preset value;

[0394] If the first data does not have a corresponding second associated data, then the intermediate ratio is determined to be the second preset value, which is greater than 1.

[0395] In some embodiments, the determining unit 903 is specifically used for:

[0396] The sum of the first probability corresponding to the first data and the second probability corresponding to the first data is determined as the target probability that the first data becomes the root cause of the alarm.

[0397] Figure 10 Schematic diagram of the data processing apparatus provided in the embodiments of this application Figure 2 .exist Figure 9 Based on what is shown, as Figure 10 As shown, the data processing device 900 also includes a processing unit 904.

[0398] The processing unit 904 is used for:

[0399] Based on the target probability of each first data point becoming the root cause of the alarm, the first data points are sorted in order of magnitude to obtain the target ranking;

[0400] Based on the target sorting, each first data point is investigated and processed in turn to eliminate alarms caused by at least one first data point.

[0401] Processing unit 904 is used for:

[0402] Obtain at least one historical data point;

[0403] Based on the generation time and data type of each historical data, at least one historical data is classified to obtain at least one historical data information. The generation time of each historical data in the historical data information is within the preset collection period, and the data type of each historical data is the same.

[0404] By performing rule-based association processing on at least one historical data information using a preset algorithm, associated data information is obtained.

[0405] It should be noted that the division of units in the embodiments of this application is illustrative and only represents one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units.

[0406] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application.

[0407] It should be noted that the apparatus provided in this application embodiment can implement all the method steps implemented by the method embodiment executed by the reader and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.

[0408] Figure 11 Schematic diagram of the data processing apparatus provided in the embodiments of this application Figure 3 ,like Figure 11 As shown, the data processing device may include: a memory 1120 for storing computer programs; a transceiver 1110 for sending and receiving data under the control of a processor 1100; and a processor 1100 for reading the computer program stored in the memory 1120 and performing the following operations:

[0409] Obtain a data processing request, which includes at least one first data point, and the generation time of the at least one data point is within a preset collection period;

[0410] Obtain related data information, which includes at least one historical data point and the relationship between every two historical data points;

[0411] Based on the data processing request and related data information, determine the target probability of each primary data point becoming the root cause of the alarm;

[0412] Among them, the first data and historical data are alarm data.

[0413] In some embodiments, the processor 1100 is configured to determine the target probability of each first data point becoming an alarm root cause based on a data processing request and associated data information, including:

[0414] In the associated data information, at least one first data information corresponding to each first data is determined;

[0415] Based on at least one piece of first data information corresponding to each first data, determine the first probability that each first data becomes the root cause of the alarm;

[0416] In the associated data information, at least one second data information corresponding to the first data is determined. The second data information includes historical data that matches at least one first data, historical data that has an association relationship with the historical data that matches at least one first data, and the association relationship between every two historical data.

[0417] Based on the second data information, determine the second probability that each of the first data points is the root cause of the alarm.

[0418] For any given first data point, determine the target probability that the first data point becomes the root cause of the alarm based on the first probability and the second probability corresponding to the first data point.

[0419] In some embodiments, the processor 1100 is configured to determine, within the associated data information, at least one first data piece of information corresponding to each first data piece of information, including:

[0420] Obtain a first quantity and a second quantity, wherein the first quantity is the total number of data included in the first data information, and the second quantity is the total number of at least one piece of the first data information;

[0421] For any given first data, determine the second data from the associated data information. The second data is historical data that matches the first data.

[0422] Based on the associated data information, the first quantity, the second quantity, and the second data, determine at least one first data information corresponding to the first data.

[0423] In some embodiments, the first quantity is M, and the second quantity is N; the processor 1100 is configured to determine at least one piece of first data information corresponding to the first data based on the associated data information, the first quantity, the second quantity, and the second data, including:

[0424] Determine the association graph corresponding to the associated data information. The association graph includes the data nodes corresponding to each historical data, the connection relationship and connection direction between each data node, and the association probability that there is an association relationship between each data node.

[0425] Based on the correlation diagram corresponding to the correlation data, determine the steady-state probability of each historical data point;

[0426] Based on M, N, and the probability of association between the data nodes, starting from the data node corresponding to the second data, and taking the reverse direction of the connection as the traversal direction, N subgraphs are obtained by traversing the association graph. Each subgraph contains M data nodes, and the starting point of the subgraph is the data node corresponding to the second data.

[0427] Based on N subgraphs, at least one first data information is determined. The first data information includes the historical data corresponding to each data node in the corresponding subgraph, and the relationship between every two historical data.

[0428] In some embodiments, the processor 1100 is configured to, for any given first data, determine the probability that the first data is an alarm root cause based on at least one first data piece of information corresponding to the first data, including:

[0429] In at least one set of first data information, determine the target quantity of the first data;

[0430] Based on the target quantity and the steady-state probability of the first data, determine the alarm root cause parameters of the first data;

[0431] The reciprocal of the alarm root cause parameter is determined as the first probability that the first data becomes the alarm root cause.

[0432] In some embodiments, the processor 1100 is configured to determine a second probability that each piece of first data is an alarm root cause based on the second data information, including:

[0433] Based on at least one first data, the target data information is determined from the second data information;

[0434] Based on the target data information, determine the second probability that each first data point is the root cause of the alarm.

[0435] In some embodiments, the processor 1100 is configured to determine target data information from second data information based on at least one first data, including:

[0436] Based on the second data information, a correlation subgraph is determined. The correlation subgraph includes at least one data node corresponding to the second data, the connection relationship and connection direction between each data node, and the correlation probability that there is a correlation relationship between each data node.

[0437] Determine whether there are independent data nodes among the data nodes;

[0438] If so, then determine at least one intermediate data corresponding to each independent data node in the second data information, add data nodes corresponding to each intermediate data in the correlation subgraph to obtain the target subgraph, and determine that the target data information includes the historical data corresponding to each data node in the target subgraph, as well as the correlation relationship between each pair of historical data.

[0439] If not, then the target data information includes the historical data corresponding to each data node in the associated subgraph, and the relationship between every two historical data.

[0440] The second data is historical data that matches at least one first data, and independent data nodes do not have any related data nodes in the association subgraph.

[0441] In some embodiments, the processor 1100 is configured to determine a second probability that each piece of first data becomes an alarm root cause based on target data information, including:

[0442] Determine the subgraph corresponding to the target data information, and the data nodes corresponding to the first data in each subgraph corresponding to the target data information;

[0443] For any first data, based on the connection direction between the data node corresponding to the first data and the adjacent data nodes, determine the first associated data node that the data node points to and the second associated data node that points to the data node in at least one adjacent data node.

[0444] The historical data corresponding to the first associated data node is determined as the first associated data corresponding to the first data, and the historical data corresponding to the second associated data node is determined as the second associated data corresponding to the first data;

[0445] Based on the first data and the first associated data corresponding to the first data, determine the first association probability between the first data and the first associated data;

[0446] Based on the first data and the second associated data corresponding to the first data, determine the second association probability between the first data and the second associated data;

[0447] Determine the steady-state probability of the first data point from the target data information;

[0448] Based on the first correlation probability corresponding to the first data, the second correlation probability corresponding to the first data, and the steady-state probability of the first data, the second probability of the first data becoming the root cause of the alarm is determined.

[0449] In some embodiments, the processor 1100 is configured to determine a second probability that the first data becomes the root cause of an alarm based on a first association probability corresponding to the first data, a second association probability corresponding to the first data, and a steady-state probability of the first data, including:

[0450] Determine the out-degree centrality of the first data based on the first association probability;

[0451] The in-degree centrality of the first data is determined based on the second association probability.

[0452] The ratio of out-degree centrality to in-degree centrality is determined as the median ratio.

[0453] Based on the median ratio and the steady-state probability corresponding to the first data, determine the second probability that the first data is the root cause of the alarm.

[0454] In some embodiments, the processor 1100 is further configured to:

[0455] If the first data does not have a corresponding first associated data, then the intermediate ratio is determined to be the first preset value;

[0456] If the first data does not have a corresponding second associated data, then the intermediate ratio is determined to be the second preset value, which is greater than 1.

[0457] In some embodiments, the processor 1100 is configured to determine the target probability of the first data becoming the root cause of an alarm based on a first probability corresponding to the first data and a second probability corresponding to the first data, including:

[0458] The sum of the first probability corresponding to the first data and the second probability corresponding to the first data is determined as the target probability that the first data becomes the root cause of the alarm.

[0459] In some embodiments, the processor 1100 is further configured to:

[0460] Based on the target probability of each first data point becoming the root cause of the alarm, the first data points are sorted in order of magnitude to obtain the target ranking;

[0461] Based on the target sorting, each first data point is investigated and processed in turn to eliminate alarms caused by at least one first data point.

[0462] In some embodiments, the processor 1100 is further configured to:

[0463] Obtain at least one historical data point;

[0464] Based on the generation time and data type of each historical data, at least one historical data is classified to obtain at least one historical data information. The generation time of each historical data in the historical data information is within the preset collection period, and the data type of each historical data is the same.

[0465] By performing rule-based association processing on at least one historical data information using a preset algorithm, associated data information is obtained.

[0466] exist Figure 11In this context, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits together, represented by one or more processors (represented by a processor) and memory (represented by memory). The bus architecture can also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides the interface. The transceiver can be multiple components, including transmitters and receivers, providing units for communicating with various other devices over transmission media, including wireless channels, wired channels, optical fibers, etc. The processor is responsible for managing the bus architecture and general processing, and the memory can store data used by the processor 1100 during operation.

[0467] The processor 1100 can be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD). The processor can also adopt a multi-core architecture.

[0468] The processor 1100 executes any of the methods provided in the embodiments of this application according to the obtained executable instructions by calling a computer program stored in the memory. The processor 1100 and the memory 1120 may also be physically separated.

[0469] It should be noted that the apparatus provided in this application embodiment can implement all the method steps implemented by the method embodiment executed by the data processing device and can achieve the same technical effect. Therefore, the parts and beneficial effects that are the same as those in the method embodiment will not be described in detail here.

[0470] This application also provides a processor-readable storage medium storing a computer program for causing the processor to execute all the method steps in the above method embodiments.

[0471] Non-transiently readable storage media can be any available medium or data storage device that the processor can access, including but not limited to magnetic storage (e.g., floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical storage (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (e.g., ROMs, EPROMs, EEPROMs, non-volatile memory (NAND flash), solid-state drives (SSDs)).

[0472] This application also provides a computer program product, including a computer program that, when executed by a processor, implements any of the methods described in the above-described method embodiments.

[0473] Processor-readable storage media can be any available medium or data storage device that the processor can access, including but not limited to magnetic storage (e.g., floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical storage (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (e.g., ROMs, EPROMs, EEPROMs, non-volatile memory (NAND flash), solid-state drives (SSDs)).

[0474] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0475] 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-executable instructions. These computer-executable 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, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0476] These processor-executable instructions may also be stored in a processor-readable memory that can instruct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the processor-readable memory produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0477] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A data processing method, characterized in that, The method includes: A data processing request is obtained, the data processing request including at least one first data, the at least one data being generated within a preset collection period; Obtain related data information, which includes at least one historical data point and the relationship between every two historical data points; Based on the data processing request and the associated data information, determine the target probability of each first data point becoming the root cause of the alarm; The first data and the historical data are alarm data.

2. The method according to claim 1, characterized in that, The step of determining the target probability of each first data point becoming an alarm root cause based on the data processing request and the associated data information includes: In the associated data information, at least one first data information corresponding to each first data is determined; Based on at least one piece of first data information corresponding to each first data, determine the first probability that each first data becomes the root cause of the alarm; In the associated data information, second data information corresponding to the at least one first data is determined. The second data information includes historical data that matches the at least one first data, historical data that has an association relationship with the historical data that matches the at least one first data, and the association relationship between every two historical data. Based on the second data information, determine the second probability that each first data point becomes the root cause of the alarm. For any given first data point, the target probability of the first data point becoming the root cause of the alarm is determined based on the first probability corresponding to the first data point and the second probability corresponding to the first data point.

3. The method according to claim 2, characterized in that, The step of determining at least one first data piece of information corresponding to each first data piece in the associated data information includes: Obtain a first quantity and a second quantity, wherein the first quantity is the total number of data included in the first data information, and the second quantity is the total number of the at least one first data information; For any given first data, a second data is determined from the associated data information, wherein the second data is historical data that matches the first data; Based on the associated data information, the first quantity, the second quantity, and the second data, at least one first data information corresponding to the first data is determined.

4. The method according to claim 3, characterized in that, The first quantity is M, and the second quantity is N; determining at least one first data piece of information corresponding to the first data based on the associated data information, the first quantity, the second quantity, and the second data includes: Determine the association graph corresponding to the associated data information. The association graph includes data nodes corresponding to each historical data, connection relationships and connection directions between each data node, and the association probability that there is an association relationship between each data node. Based on the association graph corresponding to the associated data information, determine the steady-state probability of each historical data point; Based on M, N, and the association probability between the data nodes, starting from the data node corresponding to the second data, and taking the reverse direction of the connection direction as the traversal direction, N subgraphs are obtained by traversing the association graph. Each subgraph includes M data nodes, and the starting point of each subgraph is the data node corresponding to the second data. Based on the N subgraphs, at least one first data information is determined, wherein the first data information includes the historical data corresponding to each data node in the corresponding subgraph, and the association relationship between every two historical data.

5. The method according to any one of claims 2-4, characterized in that, For any given first data, based on at least one first data piece of information corresponding to the first data, determine the first probability that the first data is the root cause of the alarm, including: In the at least one first data information, determine the target quantity of the first data; Based on the target quantity and the steady-state probability of the first data, determine the alarm root cause parameters of the first data; The reciprocal of the alarm root cause parameter is determined as the first probability that the first data becomes the alarm root cause.

6. The method according to claim 2, characterized in that, The step of determining the second probability that each piece of first data becomes the root cause of the alarm based on the second data information includes: Based on the at least one first data, target data information is determined from the second data information; Based on the target data information, determine the second probability that each first data point becomes the root cause of the alarm.

7. The method according to claim 6, characterized in that, The step of determining the target data information from the second data information based on the at least one first data includes: Based on the second data information, a correlation subgraph is determined. The correlation subgraph includes at least one data node corresponding to the second data, the connection relationship and connection direction between each data node, and the correlation probability that there is a correlation relationship between each data node. Determine whether there are independent data nodes among the data nodes; If so, then at least one intermediate data corresponding to each independent data node is determined in the second data information, and data nodes corresponding to each intermediate data are added to the associated subgraph to obtain the target subgraph, and the target data information is determined to include the historical data corresponding to each data node in the target subgraph, and the association relationship between each pair of historical data. If not, then the target data information is determined to include the historical data corresponding to each data node in the associated subgraph, and the association relationship between every two historical data. Wherein, the second data is historical data that matches the at least one first data, and the independent data node does not have any associated data nodes in the associated subgraph.

8. The method according to claim 6 or 7, characterized in that, The step of determining the second probability that each piece of first data becomes the root cause of the alarm based on the target data information includes: Determine the subgraph corresponding to the target data information, and the data nodes corresponding to each first data in the subgraph corresponding to the target data information; For any first data, based on the connection direction between the data node corresponding to the first data and the adjacent data nodes, determine the first associated data node to which the data node points and the second associated data node to which the data node points among the at least one adjacent data node; The historical data corresponding to the first associated data node is determined as the first associated data corresponding to the first data, and the historical data corresponding to the second associated data node is determined as the second associated data corresponding to the first data; Based on the first data and the first associated data corresponding to the first data, determine the first association probability between the first data and the first associated data; Based on the first data and the second associated data corresponding to the first data, determine the second association probability between the first data and the second associated data; From the target data information, determine the steady-state probability of the first data; Based on the first association probability corresponding to the first data, the second association probability corresponding to the first data, and the steady-state probability of the first data, the second probability of the first data becoming the root cause of the alarm is determined.

9. The method according to claim 8, characterized in that, The step of determining the second probability that the first data becomes the root cause of the alarm based on the first association probability corresponding to the first data, the second association probability corresponding to the first data, and the steady-state probability of the first data includes: Based on the first association probability, determine the out-degree centrality of the first data; The in-degree centrality of the first data is determined based on the second association probability. The ratio of the out-degree centrality to the in-degree centrality is determined as the median ratio. Based on the intermediate ratio and the steady-state probability corresponding to the first data, a second probability is determined that the first data is the root cause of the alarm.

10. The method according to claim 9, characterized in that, The method further includes: If the first data does not have a corresponding first associated data, then the intermediate ratio is determined to be a first preset value; If the first data does not have a corresponding second associated data, then the intermediate ratio is determined to be a second preset value, which is greater than 1.

11. The method according to claim 2, characterized in that, Determining the target probability of the first data becoming the root cause of the alarm based on the first probability corresponding to the first data and the second probability corresponding to the first data includes: The sum of the first probability corresponding to the first data and the second probability corresponding to the first data is determined as the target probability that the first data becomes the root cause of the alarm.

12. The method according to claim 1, characterized in that, The method further includes: Based on the target probability of each first data point becoming the root cause of the alarm, the first data points are sorted in order of magnitude to obtain the target ranking; According to the target sorting, each first data is checked and processed in turn to eliminate the alarm caused by at least one first data.

13. The method according to claim 1, characterized in that, The method further includes: Obtain at least one historical data point; Based on the generation time and data type of each historical data, the at least one historical data is classified to obtain at least one historical data information. The generation time of each historical data in the historical data information is within a preset collection period, and the data type of each historical data is the same. The associated data information is obtained by performing rule-based association processing on the at least one historical data information through a preset algorithm.

14. A data processing apparatus, characterized in that, include: The first acquisition unit is used to acquire a data processing request, the data processing request including at least one first data, the at least one data being generated within a preset acquisition period; The second acquisition unit is used to acquire associated data information, which includes at least one historical data and the association relationship between every two historical data. The determining unit is configured to determine the target probability of each first data becoming an alarm root cause based on the data processing request and the associated data information; The first data and the historical data are alarm data.

15. A data processing apparatus, characterized in that, Includes memory, transceiver, and processor: A memory for storing computer programs; a transceiver for sending and receiving data under the control of the processor; and a processor for reading the computer programs from the memory and performing the following operations: A data processing request is obtained, the data processing request including at least one first data, the at least one data being generated within a preset collection period; Obtain related data information, which includes at least one historical data point and the relationship between every two historical data points; Based on the data processing request and the associated data information, determine the target probability of each first data point becoming the root cause of the alarm; The first data and the historical data are alarm data.

16. The apparatus according to claim 15, characterized in that, The processor is used to determine the target probability of each first data point becoming an alarm root cause based on the data processing request and the associated data information, including: In the associated data information, at least one first data information corresponding to each first data is determined; Based on at least one piece of first data information corresponding to each first data, determine the first probability that each first data becomes the root cause of the alarm; In the associated data information, second data information corresponding to the at least one first data is determined. The second data information includes historical data that matches the at least one first data, historical data that has an association relationship with the historical data that matches the at least one first data, and the association relationship between every two historical data. Based on the second data information, determine the second probability that each first data point becomes the root cause of the alarm. For any given first data point, the target probability of the first data point becoming the root cause of the alarm is determined based on the first probability corresponding to the first data point and the second probability corresponding to the first data point.

17. The apparatus according to claim 16, characterized in that, The processor is used to determine, from the associated data information, at least one first data piece of information corresponding to each first data piece, including: Obtain a first quantity and a second quantity, wherein the first quantity is the total number of data included in the first data information, and the second quantity is the total number of the at least one first data information; For any given first data, a second data is determined from the associated data information, wherein the second data is historical data that matches the first data; Based on the associated data information, the first quantity, the second quantity, and the second data, at least one first data information corresponding to the first data is determined.

18. The apparatus according to claim 17, characterized in that, The first quantity is M, and the second quantity is N; the processor is configured to determine at least one first data piece of information corresponding to the first data based on the associated data information, the first quantity, the second quantity, and the second data, including: Determine the association graph corresponding to the associated data information. The association graph includes data nodes corresponding to each historical data, connection relationships and connection directions between each data node, and the association probability that there is an association relationship between each data node. Based on the association graph corresponding to the associated data information, determine the steady-state probability of each historical data point; Based on M, N, and the association probability between the data nodes, starting from the data node corresponding to the second data, and taking the reverse direction of the connection direction as the traversal direction, N subgraphs are obtained by traversing the association graph. Each subgraph includes M data nodes, and the starting point of each subgraph is the data node corresponding to the second data. Based on the N subgraphs, at least one first data information is determined, wherein the first data information includes the historical data corresponding to each data node in the corresponding subgraph, and the association relationship between every two historical data.

19. The apparatus according to any one of claims 16-18, characterized in that, The processor is configured to, for any given first data, determine the probability that the first data is an alarm root cause based on at least one piece of first data information corresponding to the first data, including: In the at least one first data information, determine the target quantity of the first data; Based on the target quantity and the steady-state probability of the first data, determine the alarm root cause parameters of the first data; The reciprocal of the alarm root cause parameter is determined as the first probability that the first data becomes the alarm root cause.

20. The apparatus according to claim 16, characterized in that, The processor is used to determine, based on the second data information, the second probability that each piece of first data becomes an alarm root cause, including: Based on the at least one first data, target data information is determined from the second data information; Based on the target data information, determine the second probability that each first data point becomes the root cause of the alarm.

21. The apparatus according to claim 20, characterized in that, The processor is configured to determine target data information from the second data information based on the at least one first data, including: Based on the second data information, a correlation subgraph is determined. The correlation subgraph includes at least one data node corresponding to the second data, the connection relationship and connection direction between each data node, and the correlation probability that there is a correlation relationship between each data node. Determine whether there are independent data nodes among the data nodes; If so, then at least one intermediate data corresponding to each independent data node is determined in the second data information, and data nodes corresponding to each intermediate data are added to the associated subgraph to obtain the target subgraph, and the target data information is determined to include the historical data corresponding to each data node in the target subgraph, and the association relationship between each pair of historical data. If not, then the target data information is determined to include the historical data corresponding to each data node in the associated subgraph, and the association relationship between every two historical data. Wherein, the second data is historical data that matches the at least one first data, and the independent data node does not have any associated data nodes in the associated subgraph.

22. The apparatus according to claim 20 or 21, characterized in that, The processor is used to determine a second probability that each piece of first data becomes an alarm root cause based on the target data information, including: Determine the subgraph corresponding to the target data information, and the data nodes corresponding to each first data in the subgraph corresponding to the target data information; For any first data, based on the connection direction between the data node corresponding to the first data and the adjacent data nodes, determine the first associated data node to which the data node points and the second associated data node to which the data node points among the at least one adjacent data node; The historical data corresponding to the first associated data node is determined as the first associated data corresponding to the first data, and the historical data corresponding to the second associated data node is determined as the second associated data corresponding to the first data; Based on the first data and the first associated data corresponding to the first data, determine the first association probability between the first data and the first associated data; Based on the first data and the second associated data corresponding to the first data, determine the second association probability between the first data and the second associated data; From the target data information, determine the steady-state probability of the first data; Based on the first association probability corresponding to the first data, the second association probability corresponding to the first data, and the steady-state probability of the first data, the second probability of the first data becoming the root cause of the alarm is determined.

23. The apparatus according to claim 22, characterized in that, The processor is configured to determine a second probability that the first data becomes the root cause of an alarm based on a first association probability corresponding to the first data, a second association probability corresponding to the first data, and a steady-state probability of the first data, including: Based on the first association probability, determine the out-degree centrality of the first data; The in-degree centrality of the first data is determined based on the second association probability. The ratio of the out-degree centrality to the in-degree centrality is determined as the median ratio. Based on the intermediate ratio and the steady-state probability corresponding to the first data, a second probability is determined that the first data is the root cause of the alarm.

24. The apparatus according to claim 23, characterized in that, The processor is also used for: If the first data does not have a corresponding first associated data, then the intermediate ratio is determined to be a first preset value; If the first data does not have a corresponding second associated data, then the intermediate ratio is determined to be a second preset value, which is greater than 1.

25. The apparatus according to claim 16, characterized in that, The processor is configured to determine the target probability of the first data becoming the root cause of an alarm based on a first probability corresponding to the first data and a second probability corresponding to the first data, including: The sum of the first probability corresponding to the first data and the second probability corresponding to the first data is determined as the target probability that the first data becomes the root cause of the alarm.

26. The apparatus according to claim 15, characterized in that, The processor is also used for: Based on the target probability of each first data point becoming the root cause of the alarm, the first data points are sorted in order of magnitude to obtain the target ranking; According to the target sorting, each first data is checked and processed in turn to eliminate the alarm caused by at least one first data.

27. The apparatus according to claim 15, characterized in that, The processor is also used for: Obtain at least one historical data point; Based on the generation time and data type of each historical data, the at least one historical data is classified to obtain at least one historical data information. The generation time of each historical data in the historical data information is within a preset collection period, and the data type of each historical data is the same. The associated data information is obtained by performing rule-based association processing on the at least one historical data information through a preset algorithm.

28. A non-transiently readable storage medium, characterized in that, The non-transiently readable storage medium stores a computer program that causes a processor to perform the method according to any one of claims 1 to 13.

Images