Improvement of multi-parameter anomaly detection method: Addition of a relational token between parameters

In the continuous development of systems, the increasing volume and complexity of data that engineers must analyze have become significant challenges. To address this issue, extensive research has been conducted on automated anomaly detection in logs. However, due to the limited variety of available...

Full description

Saved in:
Bibliographic Details
Main Authors: Hironori Uchida, Keitaro Tominaga, Hideki Itai, Yujie Li, Yoshihisa Nakatoh
Format: Article
Language:English
Published: KeAi Communications Co. Ltd. 2025-01-01
Series:Cognitive Robotics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2667241325000114
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849711738550747136
author Hironori Uchida
Keitaro Tominaga
Hideki Itai
Yujie Li
Yoshihisa Nakatoh
author_facet Hironori Uchida
Keitaro Tominaga
Hideki Itai
Yujie Li
Yoshihisa Nakatoh
author_sort Hironori Uchida
collection DOAJ
description In the continuous development of systems, the increasing volume and complexity of data that engineers must analyze have become significant challenges. To address this issue, extensive research has been conducted on automated anomaly detection in logs. However, due to the limited variety of available datasets, most studies have focused on sequence-based anomalies in logs, with relatively little attention paid to parameter-based anomaly detection. To bridge this gap, we prepared a labeled dataset specifically designed for parameter-based anomaly detection and propose a novel method utilizing BERTMaskedLM. Since continuously changing logs in system development are difficult to label, we also propose a method that enables learning without labeled data. Previous studies have employed BERTMaskedLM to capture relationships between parameters in multi-parameter logs for anomaly detection. However, a known issue arises when the ranges of numerical parameters overlap, resulting in reduced detection accuracy. To mitigate this, we introduced tokens that encode the relationships between parameters, improving the independence of parameter combinations and enhancing anomaly detection accuracy (increasing the F1-score by more than 0.002). In this study, we employed a simple yet effective approach by using the total value of each token as the added token. Since only the parameter portions vary within the same log template structure, these proposed tokens effectively capture the relationships between parameters. Additionally, we visualized the influence of the added tokens and conducted experiments using a new dataset to assess the reliability of our proposed method.
format Article
id doaj-art-ee9e377f5b8a4399909fb09aaad9d767
institution DOAJ
issn 2667-2413
language English
publishDate 2025-01-01
publisher KeAi Communications Co. Ltd.
record_format Article
series Cognitive Robotics
spelling doaj-art-ee9e377f5b8a4399909fb09aaad9d7672025-08-20T03:14:32ZengKeAi Communications Co. Ltd.Cognitive Robotics2667-24132025-01-01517619110.1016/j.cogr.2025.03.004Improvement of multi-parameter anomaly detection method: Addition of a relational token between parametersHironori Uchida0Keitaro Tominaga1Hideki Itai2Yujie Li3Yoshihisa Nakatoh4Kyushu Institute of Technology, Kitakyushu, Japan; Corresponding author.Panasonic System Design Co., Ltd, Yokohama, JapanPanasonic System Design Co., Ltd, Yokohama, JapanKyushu Institute of Technology, Kitakyushu, JapanKyushu Institute of Technology, Kitakyushu, JapanIn the continuous development of systems, the increasing volume and complexity of data that engineers must analyze have become significant challenges. To address this issue, extensive research has been conducted on automated anomaly detection in logs. However, due to the limited variety of available datasets, most studies have focused on sequence-based anomalies in logs, with relatively little attention paid to parameter-based anomaly detection. To bridge this gap, we prepared a labeled dataset specifically designed for parameter-based anomaly detection and propose a novel method utilizing BERTMaskedLM. Since continuously changing logs in system development are difficult to label, we also propose a method that enables learning without labeled data. Previous studies have employed BERTMaskedLM to capture relationships between parameters in multi-parameter logs for anomaly detection. However, a known issue arises when the ranges of numerical parameters overlap, resulting in reduced detection accuracy. To mitigate this, we introduced tokens that encode the relationships between parameters, improving the independence of parameter combinations and enhancing anomaly detection accuracy (increasing the F1-score by more than 0.002). In this study, we employed a simple yet effective approach by using the total value of each token as the added token. Since only the parameter portions vary within the same log template structure, these proposed tokens effectively capture the relationships between parameters. Additionally, we visualized the influence of the added tokens and conducted experiments using a new dataset to assess the reliability of our proposed method.http://www.sciencedirect.com/science/article/pii/S2667241325000114Log anomaly detectionParameter anomaly detectionTransformerMulti parameterSoftware log
spellingShingle Hironori Uchida
Keitaro Tominaga
Hideki Itai
Yujie Li
Yoshihisa Nakatoh
Improvement of multi-parameter anomaly detection method: Addition of a relational token between parameters
Cognitive Robotics
Log anomaly detection
Parameter anomaly detection
Transformer
Multi parameter
Software log
title Improvement of multi-parameter anomaly detection method: Addition of a relational token between parameters
title_full Improvement of multi-parameter anomaly detection method: Addition of a relational token between parameters
title_fullStr Improvement of multi-parameter anomaly detection method: Addition of a relational token between parameters
title_full_unstemmed Improvement of multi-parameter anomaly detection method: Addition of a relational token between parameters
title_short Improvement of multi-parameter anomaly detection method: Addition of a relational token between parameters
title_sort improvement of multi parameter anomaly detection method addition of a relational token between parameters
topic Log anomaly detection
Parameter anomaly detection
Transformer
Multi parameter
Software log
url http://www.sciencedirect.com/science/article/pii/S2667241325000114
work_keys_str_mv AT hironoriuchida improvementofmultiparameteranomalydetectionmethodadditionofarelationaltokenbetweenparameters
AT keitarotominaga improvementofmultiparameteranomalydetectionmethodadditionofarelationaltokenbetweenparameters
AT hidekiitai improvementofmultiparameteranomalydetectionmethodadditionofarelationaltokenbetweenparameters
AT yujieli improvementofmultiparameteranomalydetectionmethodadditionofarelationaltokenbetweenparameters
AT yoshihisanakatoh improvementofmultiparameteranomalydetectionmethodadditionofarelationaltokenbetweenparameters