Fault log classification method, system and device and medium

Abstract

The invention discloses a fault log classification method, which comprises the following steps: receiving fault logs to be classified, and determining a plurality of segmentation positions corresponding to the fault logs to be classified according to phrases containing most vocabularies in a preset corpus; segmenting the to-be-classified fault logs according to the corresponding segmentation positions to obtain a plurality of word groups; determining the weight of each phrase according to the corpus, and screening out a plurality of phrases according to the weights; and calculating the similarity between the to-be-classified fault logs and each classified fault log by utilizing the plurality of word groups screened by the plurality of classified fault logs according to the weights and theplurality of word groups screened by the to-be-classified fault logs according to the weights, and further classifying the to-be-classified fault logs according to the similarity. The invention further discloses a system, computer equipment and a readable storage medium. According to the scheme provided by the invention, the English logs are processed into phrases and phrases, so that the vocabulary amount of subsequent processing is greatly reduced.

Description

technical field [0001] The field of log processing of the present invention specifically relates to a fault log classification method, system, device and storage medium. Background technique [0002] In server monitoring technology, it is a very common and effective solution to predict and locate faults through server daily operation log analysis. Log files provide a large amount of information, and many algorithms have emerged from this, such as analyzing log text, extracting text features, and establishing Text feature model to classify text is a common technical solution at present. However, since the server log data is in English, each word in the English text has spaces, and using the text feature extraction scheme to extract each key word will result in too large a volume of data, too high a feature vector dimension, and a large amount of calculation. Contents of the invention [0003] In view of this, in order to overcome at least one aspect of the above problems, ...

AI Technical Summary

Problems solved by technology

However, since the server log data is in English, and each word in the English text has spaces, using the text feature extraction scheme to extract each key word will lead to excessive data volume, high feature vector dimensions, and a large amount of calculation.

Images