Search

CN-122020197-A - Operation and maintenance knowledge graph matching method, device, equipment and storage medium

CN122020197ACN 122020197 ACN122020197 ACN 122020197ACN-122020197-A

Abstract

The invention discloses an operation and maintenance knowledge graph matching method, device, equipment and storage medium, comprising the steps of obtaining historical fault data and constructing a fault knowledge graph according to the historical fault data; when a new fault occurs, extracting new fault characteristics, matching the new fault characteristics through a fault knowledge graph to obtain similar fault data, and generating a fault matching report based on the similar fault data. Information isolation can be broken through by constructing a fault knowledge graph, multidimensional association of fault related entities is established, and hidden causal relationships are mined. The core information of the new fault can be rapidly extracted by extracting the new fault characteristics, the subsequent matching efficiency is improved, the highly-matched historical cases can be rapidly positioned by matching the new fault characteristics through the knowledge graph, the accurate and comprehensive matching result is ensured, and the manual screening cost is reduced. The generated fault matching report can structurally present similar fault key information, and provides investigation direction and action suggestions for users intuitively, so that decision difficulty is reduced.

Inventors

  • ZHANG YATING

Assignees

  • 北京优特捷信息技术有限公司

Dates

Publication Date
20260512
Application Date
20260126

Claims (10)

  1. 1. The operation and maintenance knowledge graph matching method is characterized by comprising the following steps of: acquiring historical fault data, and constructing a fault knowledge graph according to the historical fault data; When a new fault occurs, extracting new fault characteristics, and matching the new fault characteristics through the fault knowledge graph to obtain similar fault data; and generating a fault matching report based on the similar fault data.
  2. 2. The method of claim 1, wherein the obtaining historical fault data comprises: defining a data acquisition range, wherein the data acquisition range comprises basic attributes, fault characteristics, quantization indexes, analysis results and solutions; Acquiring the solved original fault data in the operation and maintenance scene based on the data acquisition range; Performing data cleaning on the original fault data to generate cleaned original fault data; And carrying out structural mapping on the cleaned original fault data according to a preset field to generate historical fault data.
  3. 3. The method of claim 2, wherein constructing a fault knowledge-graph from the historical fault data comprises: Extracting map nodes from historical fault data according to predefined entity types, wherein the entity types comprise service types, resource types, error code types, index types, root cause types and operation types; establishing semantic associations in the extracted graph nodes based on predefined relationship types, forming graph links, wherein the relationship types comprise generation, causing, influencing, aggravating and solving; and storing the identified map nodes and the established map links into a map database to form a fault knowledge map.
  4. 4. The method of claim 1, wherein extracting new fault signatures when a new fault occurs comprises: when a new fault occurs, collecting new fault data in real time, wherein the new fault data comprises alarm data, log data and monitoring data; Extracting initial fault characteristics from the new fault data according to preset dimensions, wherein the preset dimensions comprise keywords, log templates, abnormal indexes, topological services and root cause categories; and carrying out integrity check on the initial fault characteristics, and forming new fault characteristics after the verification is passed.
  5. 5. The method according to claim 1, wherein said matching the new fault signature by the fault knowledge graph to obtain similar fault data comprises: searching in a fault knowledge graph directly based on the keywords in the new fault characteristics to obtain direct matching fault data; performing association expansion retrieval in the fault knowledge graph based on the new fault characteristics to obtain indirect matching fault data; and determining similar fault data according to the direct matching fault data and the indirect matching fault data.
  6. 6. The method of claim 5, wherein said determining similar fault data from said direct match fault data and said indirect match fault data comprises: Combining the direct matching fault and the indirect matching fault data, and removing repeated faults to form a candidate fault set; calculating the similarity between each candidate fault data in the candidate fault set and the new fault characteristic to obtain a similarity score; sequencing the candidate fault data according to the sequence of the similarity score from high to low to obtain a sequenced candidate fault set; and selecting a designated number of candidate fault data from the sorted candidate fault sets to serve as similar fault data.
  7. 7. The method according to claim 1, characterized in that the method further comprises: Acquiring user feedback information based on the fault matching report, wherein the user feedback information comprises effective information and ineffective information; And updating the fault knowledge graph based on the effective information.
  8. 8. An operation and maintenance knowledge graph matching device is characterized by comprising: the knowledge graph construction module is used for acquiring historical fault data and constructing a fault knowledge graph according to the historical fault data; The similar fault matching module is used for extracting new fault characteristics when new faults occur, and matching the new fault characteristics through the fault knowledge graph to obtain similar fault data; and the matching report generation module is used for generating a fault matching report based on the similar fault data.
  9. 9. An electronic device, the electronic device comprising: At least one processor; and a memory communicatively coupled to the at least one processor; Wherein the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.
  10. 10. A computer storage medium storing computer instructions for causing a processor to perform the method of any one of claims 1-7 when executed.

Description

Operation and maintenance knowledge graph matching method, device, equipment and storage medium Technical Field The present invention relates to the field of system operation and maintenance, and in particular, to an operation and maintenance knowledge graph matching method, apparatus, device, and storage medium. Background In a digital operation and maintenance scene, the complexity of the system is continuously improved, fault trigger factors are increasingly diversified, and multi-source heterogeneous data such as logs, monitoring indexes, service topology and the like are in explosive growth. Operation and maintenance fault diagnosis directly relates to service continuity and system availability, and rapidly positioning root causes and multiplexing history processing experience become core requirements. The prior art in the field of operation and maintenance fault diagnosis mainly comprises two categories, namely a historical fault storage and retrieval scheme based on a traditional database, a basic information, a root cause and a solution of a fault are stored in a form, keywords are relied on to be matched accurately during retrieval, and a fault correlation analysis technology based on a simple algorithm is used for carrying out preliminary filtration and feature extraction on fault data to assist operation and maintenance personnel to locate fault points. The traditional database storage mode leads to isolated fault related knowledge, indirect association among service, indexes and root causes cannot be found, a simple algorithm can only complete data preliminary processing, effective integration of historical experience is lacked, fault diagnosis still highly depends on personal experience of senior operation and maintenance personnel, experience is difficult to inherit, new people process fault threshold is high, meanwhile, the prior art does not solve the problem of deep fusion of multi-source heterogeneous data, cannot carry out accurate matching based on comprehensive features, so that the average time of fault location is tens of hours, diagnosis efficiency is low, partial faults even cannot find definite root cause references, and service recovery speed is seriously influenced. Disclosure of Invention The invention provides an operation and maintenance knowledge graph matching method, an operation and maintenance knowledge graph matching device, equipment and a storage medium, which solve the technical problems that the conventional operation and maintenance fault diagnosis depends on personal experience, knowledge is difficult to inherit, diagnosis efficiency is low and multi-source heterogeneous data is difficult to effectively fuse by constructing an operation and maintenance knowledge graph fused with fault multi-dimensional characteristics and adopting a multi-dimensional weighted similarity intelligent matching method. According to an aspect of the present invention, there is provided an operation and maintenance knowledge graph matching method, including: Acquiring historical fault data, and constructing a fault knowledge graph according to the historical fault data; When a new fault occurs, extracting new fault characteristics, and matching the new fault characteristics through a fault knowledge graph to obtain similar fault data; a fault-matching report is generated based on the similar fault data. The method comprises the steps of acquiring historical fault data, namely defining a data acquisition range, wherein the data acquisition range comprises basic attributes, fault characteristics, quantization indexes, analysis results and solutions, acquiring original fault data which are solved in an operation and maintenance scene based on the data acquisition range, performing data cleaning on the original fault data to generate cleaned original fault data, and performing structural mapping on the cleaned original fault data according to preset fields to generate the historical fault data. Optionally, constructing a fault knowledge graph according to the historical fault data comprises extracting graph nodes from the historical fault data according to a predefined entity type, wherein the entity type comprises a service type, a resource type, an error code type, an index type, a root cause type and an operation type, establishing semantic association in the extracted graph nodes based on a predefined relation type to form a graph link, wherein the relation type comprises generation, causing, influence, aggravation and solution, and storing the identified graph nodes and the established graph link into a graph database to form the fault knowledge graph. Optionally, when a new fault occurs, extracting new fault characteristics comprises the steps of collecting new fault data in real time when the new fault occurs, wherein the new fault data comprise alarm data, log data and monitoring data, extracting initial fault characteristics from the new fault data according to preset dimensions, wh