LOG MESSAGE GROUPING APPARATUS, LOG MESSAGE GROUPING SYSTEM, AND LOG MESSAGE GROUPING METHOD
First Claim
1. A log message grouping apparatus comprising a processor, the log message grouping apparatus being characterized in that the processor is configured to(a) collect a log that records a plurality of log messages,(b) acquire the plurality of log messages from the log,(c) acquire an occurrence frequency of each word that forms a first log message from word-occurrence frequency association information that is created based on the plurality of log messages contained in the log and stores each word contained in the plurality of log messages and the occurrence frequency of the word with the word and the occurrence frequency associated with each other,(d) store, in log-word association information, the first log message and a word that is one of words that form the first log message and occurs by a frequency smaller than a predetermined occurrence frequency determination threshold with the first log message and the word as a minority word associated with each other,(e) acquire the minority word contained in a second log message earlier than the first log message from the log-word association information, and(f) calculate a first evaluation value that is an evaluation value representing a degree of coincidence between the minority word contained in the first log message and the minority word contained in the second log message.
1 Assignment
0 Petitions
Accused Products
Abstract
A log message grouping apparatus calculates a coincidence degree evaluation value E1 representing the degree of coincidence between minority words, which are relatively low occurrence frequency words, out of the words that form a log message and minority words in another log message. The log message grouping apparatus further calculates an occurrence degree evaluation value E2 representing the degree of occurrence of the combination of a classification value of a log message and the classification value of a past log message on the basis of the occurrence frequency of the combination of the classification values and calculates an association degree evaluation value E3 representing the degree of association between the log message and the past log message on the basis of the coincidence degree evaluation value E1 and the occurrence degree evaluation value E2.
-
Citations
9 Claims
-
1. A log message grouping apparatus comprising a processor, the log message grouping apparatus being characterized in that the processor is configured to
(a) collect a log that records a plurality of log messages, (b) acquire the plurality of log messages from the log, (c) acquire an occurrence frequency of each word that forms a first log message from word-occurrence frequency association information that is created based on the plurality of log messages contained in the log and stores each word contained in the plurality of log messages and the occurrence frequency of the word with the word and the occurrence frequency associated with each other, (d) store, in log-word association information, the first log message and a word that is one of words that form the first log message and occurs by a frequency smaller than a predetermined occurrence frequency determination threshold with the first log message and the word as a minority word associated with each other, (e) acquire the minority word contained in a second log message earlier than the first log message from the log-word association information, and (f) calculate a first evaluation value that is an evaluation value representing a degree of coincidence between the minority word contained in the first log message and the minority word contained in the second log message.
-
9. A log message grouping method, characterized in that
a processor is configured to (a) collect a log that records a plurality of log messages, (b) acquire the plurality of log messages from the log, (c) acquire an occurrence frequency of each word that forms a first log message from word-occurrence frequency association information that is created based on the plurality of log messages contained in the log and stores each word contained in the plurality of log messages and the occurrence frequency of the word with the word and the occurrence frequency associated with each other, (d) store, in log-word association information, the first log message and a word that is one of words that form the first log message and occurs by a frequency smaller than a predetermined occurrence frequency determination threshold with the first log message and the word as a minority word associated with each other, (e) acquire the minority word contained in a second log message earlier than the first log message from the log-word association information, and (f) calculate a first evaluation value that is an evaluation value representing a degree of coincidence between the minority word contained in the first log message and the minority word contained in the second log message.
Specification