Log template mining method and system based on multiple indexes

Through the multi-index log template mining method, log items are preprocessed and classified, and multiple index data are used to evaluate and adjust index combinations, the problem of lack of universality and low efficiency of log template mining in the existing technology is solved, and log template mining with high accuracy and high efficiency is achieved.

CN119691694AActive Publication Date: 2025-03-25Chinese People's Liberation Army Cyberspace Force Information Engineering University
7 Cites 0 Cited by

Patent Information

Application Number
CN202411760771.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-03
Publication Date
2025-03-25
Estimated Expiration
2044-12-03

AI Technical Summary

Technical Problem

The existing log template mining methods lack universality and low mining efficiency, making it difficult to find frequent occurrences of events worthy of attention.

Method used

A multi-index-based log template mining method is adopted to preprocess the original log data, extract the log item set, and form the initial log template based on the sample log items. Use multiple index data (log item length, first word, punctuation, keyword) to evaluate the initial log template differentiation, select the target index combination, and classify the log items according to the index combination until an intermediate log template that meets expectations is generated.

Benefits of technology

It improves the accuracy and efficiency of log template mining, adapts to the mining tasks of multiple log templates, reduces manual intervention, and realizes adaptive index combination recommendations.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of log information processing, in particular to a multi-index-based log template mining method and system.The collected original log data is preprocessed, a log item set in a log is extracted, and the log item set comprises key events and key variables; sampling the extracted log item set to obtain sampling log items, and forming a plurality of initial log templates based on the sampling log items; evaluating the distinction degree of the initial log template by using the index data, selecting a target index according to the distinction degree and the performance overhead, and forming an index combination; and classifying the log items according to the index combination, dividing the log items into a plurality of clusters and generating an intermediate log template, and evaluating the distinction degree of the intermediate log template by utilizing the index combination so as to adaptively adjust the index combination and realize log template mining. On the basis of reducing manual intervention, the method is suitable for mining tasks of various log templates and has high universality.
Need to check novelty before this filing date? Find Prior Art