Presentation 2010-07-09
Trouble Cause Estimation Method from Past Cases in Incident Management
Kuniaki SHIMADA, Yasuhide MATSUMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The maintenance cost for application and operation in the IT investment cost has exceed 70%. It is necessary to identify the trouble cause in incident management promptly to reduce this cost. In a past method, an important word extraction and the machine learning like the decision tree is used to estimate the trouble cause. In this method, there is a problem with low identification rate and accuracy when the error message is not described in a past case. Then, we propose the method for raising an identification rate and accuracy of the trouble cause. In our method, evidence information is regularized and input to the incident as tag. An identification rate has improved to 63% as a result of actually evaluation by the incident of the site where the application is maintained.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Operation Management / ITIL / Incident Management / Machine Learning / Decision Tree
Paper # ICM2010-22
Date of Issue

Conference Information
Committee ICM
Conference Date 2010/7/1(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Information and Communication Management(ICM)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Trouble Cause Estimation Method from Past Cases in Incident Management
Sub Title (in English)
Keyword(1) Operation Management
Keyword(2) ITIL
Keyword(3) Incident Management
Keyword(4) Machine Learning
Keyword(5) Decision Tree
1st Author's Name Kuniaki SHIMADA
1st Author's Affiliation Research Center for Cloud Computing, Fujitsu Laboratories Ltd.()
2nd Author's Name Yasuhide MATSUMOTO
2nd Author's Affiliation Research Center for Cloud Computing, Fujitsu Laboratories Ltd.
Date 2010-07-09
Paper # ICM2010-22
Volume (vol) vol.110
Number (no) 119
Page pp.pp.-
#Pages 6
Date of Issue