TY - GEN
T1 - An adaptive abnormal behavior detection using online sequential learning
AU - Ito, Rei
AU - Tsukada, Mineto
AU - Kondo, Masaaki
AU - Matsutani, Hiroki
N1 - Funding Information:
Acknowledgements This work was supported by JST CREST Grant Number JPMJCR1785, Japan.
Publisher Copyright:
© 2019 IEEE.
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.
T2 - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019
Y2 - 1 August 2019 through 3 August 2019
ER -