An adaptive abnormal behavior detection using online sequential learning

Rei Ito, Mineto Tsukada, Masaaki Kondo, Hiroki Matsutani

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the real world, normal and abnormal behavior patterns vary depending on a given environment, which means that the abnormal behavior detection model should be customized. To address this issue, in this paper, we employ OS-ELM (Online Sequential Extreme Learning Machine) and Autoencoder for adaptive abnormal behavior detection. First, state-transition probability tables of a target during an initial learning period are learned as normal behaviors. Then, Autoencoder-based anomaly detection is performed for the state-transition probability tables of subsequent time frames. The abnormal behavior detection model is updated by using OS-ELM algorithm every time a new probability table or behavior comes. The number of abnormal behavior detection instances is dynamically tuned to reflect the recent normal patterns or modes. Also, the table is compressed to reduce the computation cost. Evaluation results using a driving dataset of cars show that the proposed abnormal behavior detection accurately identifies normal and aggressive driving patterns with the optimal number of the abnormal behavior detection instances.

Original languageEnglish
Title of host publicationProceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019
EditorsMeikang Qiu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages436-440
Number of pages5
ISBN (Electronic)9781728116631
DOIs
Publication statusPublished - 2019 Aug
Event22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019 - New York, United States
Duration: 2019 Aug 12019 Aug 3

Publication series

NameProceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019

Conference

Conference22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019
CountryUnited States
CityNew York
Period19/8/119/8/3

Fingerprint

Learning systems
Railroad cars
Costs

Keywords

  • Abnormal behavior detection
  • Autoencoder
  • Clustering
  • Machine learning
  • OS ELM

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Human-Computer Interaction
  • Artificial Intelligence

Cite this

Ito, R., Tsukada, M., Kondo, M., & Matsutani, H. (2019). An adaptive abnormal behavior detection using online sequential learning. In M. Qiu (Ed.), Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019 (pp. 436-440). [8919512] (Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CSE/EUC.2019.00087

An adaptive abnormal behavior detection using online sequential learning. / Ito, Rei; Tsukada, Mineto; Kondo, Masaaki; Matsutani, Hiroki.

Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019. ed. / Meikang Qiu. Institute of Electrical and Electronics Engineers Inc., 2019. p. 436-440 8919512 (Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ito, R, Tsukada, M, Kondo, M & Matsutani, H 2019, An adaptive abnormal behavior detection using online sequential learning. in M Qiu (ed.), Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019., 8919512, Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019, Institute of Electrical and Electronics Engineers Inc., pp. 436-440, 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019, New York, United States, 19/8/1. https://doi.org/10.1109/CSE/EUC.2019.00087
Ito R, Tsukada M, Kondo M, Matsutani H. An adaptive abnormal behavior detection using online sequential learning. In Qiu M, editor, Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 436-440. 8919512. (Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019). https://doi.org/10.1109/CSE/EUC.2019.00087
Ito, Rei ; Tsukada, Mineto ; Kondo, Masaaki ; Matsutani, Hiroki. / An adaptive abnormal behavior detection using online sequential learning. Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019. editor / Meikang Qiu. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 436-440 (Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019).
@inproceedings{5821f675764243f28c9c6b7fe282b6f5,
title = "An adaptive abnormal behavior detection using online sequential learning",
abstract = "In the real world, normal and abnormal behavior patterns vary depending on a given environment, which means that the abnormal behavior detection model should be customized. To address this issue, in this paper, we employ OS-ELM (Online Sequential Extreme Learning Machine) and Autoencoder for adaptive abnormal behavior detection. First, state-transition probability tables of a target during an initial learning period are learned as normal behaviors. Then, Autoencoder-based anomaly detection is performed for the state-transition probability tables of subsequent time frames. The abnormal behavior detection model is updated by using OS-ELM algorithm every time a new probability table or behavior comes. The number of abnormal behavior detection instances is dynamically tuned to reflect the recent normal patterns or modes. Also, the table is compressed to reduce the computation cost. Evaluation results using a driving dataset of cars show that the proposed abnormal behavior detection accurately identifies normal and aggressive driving patterns with the optimal number of the abnormal behavior detection instances.",
keywords = "Abnormal behavior detection, Autoencoder, Clustering, Machine learning, OS ELM",
author = "Rei Ito and Mineto Tsukada and Masaaki Kondo and Hiroki Matsutani",
year = "2019",
month = "8",
doi = "10.1109/CSE/EUC.2019.00087",
language = "English",
series = "Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "436--440",
editor = "Meikang Qiu",
booktitle = "Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019",

}

TY - GEN

T1 - An adaptive abnormal behavior detection using online sequential learning

AU - Ito, Rei

AU - Tsukada, Mineto

AU - Kondo, Masaaki

AU - Matsutani, Hiroki

PY - 2019/8

Y1 - 2019/8

N2 - In the real world, normal and abnormal behavior patterns vary depending on a given environment, which means that the abnormal behavior detection model should be customized. To address this issue, in this paper, we employ OS-ELM (Online Sequential Extreme Learning Machine) and Autoencoder for adaptive abnormal behavior detection. First, state-transition probability tables of a target during an initial learning period are learned as normal behaviors. Then, Autoencoder-based anomaly detection is performed for the state-transition probability tables of subsequent time frames. The abnormal behavior detection model is updated by using OS-ELM algorithm every time a new probability table or behavior comes. The number of abnormal behavior detection instances is dynamically tuned to reflect the recent normal patterns or modes. Also, the table is compressed to reduce the computation cost. Evaluation results using a driving dataset of cars show that the proposed abnormal behavior detection accurately identifies normal and aggressive driving patterns with the optimal number of the abnormal behavior detection instances.

AB - In the real world, normal and abnormal behavior patterns vary depending on a given environment, which means that the abnormal behavior detection model should be customized. To address this issue, in this paper, we employ OS-ELM (Online Sequential Extreme Learning Machine) and Autoencoder for adaptive abnormal behavior detection. First, state-transition probability tables of a target during an initial learning period are learned as normal behaviors. Then, Autoencoder-based anomaly detection is performed for the state-transition probability tables of subsequent time frames. The abnormal behavior detection model is updated by using OS-ELM algorithm every time a new probability table or behavior comes. The number of abnormal behavior detection instances is dynamically tuned to reflect the recent normal patterns or modes. Also, the table is compressed to reduce the computation cost. Evaluation results using a driving dataset of cars show that the proposed abnormal behavior detection accurately identifies normal and aggressive driving patterns with the optimal number of the abnormal behavior detection instances.

KW - Abnormal behavior detection

KW - Autoencoder

KW - Clustering

KW - Machine learning

KW - OS ELM

UR - http://www.scopus.com/inward/record.url?scp=85077081004&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077081004&partnerID=8YFLogxK

U2 - 10.1109/CSE/EUC.2019.00087

DO - 10.1109/CSE/EUC.2019.00087

M3 - Conference contribution

AN - SCOPUS:85077081004

T3 - Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019

SP - 436

EP - 440

BT - Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019

A2 - Qiu, Meikang

PB - Institute of Electrical and Electronics Engineers Inc.

ER -