Diagnosis of Stochastic Discrete Event Systems Based on <I>N</I>-gram Models

Miwa YOSHIMOTO; Koichi KOBAYASHI; Kunihiko HIRAISHI

doi:10.1587/transfun.E98.A.618

Diagnosis of Stochastic Discrete Event Systems Based on N-gram Models

Miwa YOSHIMOTO, Koichi KOBAYASHI, Kunihiko HIRAISHI

Full Text Views

0

Share
Cite this

Summary :

In this paper, we present a new method for diagnosis of stochastic discrete event system. The method is based on anomaly detection for sequences. We call the method sequence profiling (SP). SP does not require any system models and any system-specific knowledge. The only information necessary for SP is event logs from the target system. Using event logs from the system in the normal situation, N-gram models are learned, where the N-gram model is used as approximation of the system behavior. Based on the N-gram model, the diagnoser estimates what kind of faults has occurred in the system, or may conclude that no faults occurs. Effectiveness of the proposed method is demonstrated by application to diagnosis of a multi-processor system.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E98-A No.2 pp.618-625

Publication Date: 2015/02/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E98.A.618

Type of Manuscript: Special Section PAPER (Special Section on Mathematical Systems Science and its Applications)

Category

Authors

Miwa YOSHIMOTO
  Japan Advanced Institute of Science and Technology
Koichi KOBAYASHI
  Japan Advanced Institute of Science and Technology
Kunihiko HIRAISHI
  Japan Advanced Institute of Science and Technology

Keyword

stochastic discrete event systems, diagnosis, N-gram models

Cite this

Copy

Miwa YOSHIMOTO, Koichi KOBAYASHI, Kunihiko HIRAISHI, "Diagnosis of Stochastic Discrete Event Systems Based on N-gram Models" in IEICE TRANSACTIONS on Fundamentals, vol. E98-A, no. 2, pp. 618-625, February 2015, doi: 10.1587/transfun.E98.A.618.
Abstract: In this paper, we present a new method for diagnosis of stochastic discrete event system. The method is based on anomaly detection for sequences. We call the method sequence profiling (SP). SP does not require any system models and any system-specific knowledge. The only information necessary for SP is event logs from the target system. Using event logs from the system in the normal situation, N-gram models are learned, where the N-gram model is used as approximation of the system behavior. Based on the N-gram model, the diagnoser estimates what kind of faults has occurred in the system, or may conclude that no faults occurs. Effectiveness of the proposed method is demonstrated by application to diagnosis of a multi-processor system.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1587/transfun.E98.A.618/_p

Copy

@ARTICLE{e98-a_2_618,
author={Miwa YOSHIMOTO, Koichi KOBAYASHI, Kunihiko HIRAISHI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Diagnosis of Stochastic Discrete Event Systems Based on N-gram Models},
year={2015},
volume={E98-A},
number={2},
pages={618-625},
abstract={In this paper, we present a new method for diagnosis of stochastic discrete event system. The method is based on anomaly detection for sequences. We call the method sequence profiling (SP). SP does not require any system models and any system-specific knowledge. The only information necessary for SP is event logs from the target system. Using event logs from the system in the normal situation, N-gram models are learned, where the N-gram model is used as approximation of the system behavior. Based on the N-gram model, the diagnoser estimates what kind of faults has occurred in the system, or may conclude that no faults occurs. Effectiveness of the proposed method is demonstrated by application to diagnosis of a multi-processor system.},
keywords={},
doi={10.1587/transfun.E98.A.618},
ISSN={1745-1337},
month={February},}

Copy

TY - JOUR
TI - Diagnosis of Stochastic Discrete Event Systems Based on N-gram Models
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 618
EP - 625
AU - Miwa YOSHIMOTO
AU - Koichi KOBAYASHI
AU - Kunihiko HIRAISHI
PY - 2015
DO - 10.1587/transfun.E98.A.618
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E98-A
IS - 2
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - February 2015
AB - In this paper, we present a new method for diagnosis of stochastic discrete event system. The method is based on anomaly detection for sequences. We call the method sequence profiling (SP). SP does not require any system models and any system-specific knowledge. The only information necessary for SP is event logs from the target system. Using event logs from the system in the normal situation, N-gram models are learned, where the N-gram model is used as approximation of the system behavior. Based on the N-gram model, the diagnoser estimates what kind of faults has occurred in the system, or may conclude that no faults occurs. Effectiveness of the proposed method is demonstrated by application to diagnosis of a multi-processor system.
ER -