Title :
Anomaly detection from multivariate time-series with sparse representation
Author :
Takeishi, Naoya ; Yairi, Takehisa
Author_Institution :
Dept. of Aeronaut. & Astronaut., Univ. of Tokyo, Tokyo, Japan
Abstract :
Anomaly detection from sensor data is an important data mining application for efficient and secure operation of complicated systems. In this study, we propose a novel anomaly detection method for multivariate time-series to capture relationships of variables and time-domain correlations simultaneously, without assuming any generative models of signals. The supposed framework in this study is a semi-supervised anomaly detection where we seek unusual parts of test data compared with reference data. The proposed method is based on feature extraction with sparse representation and relationship learning with dimensionality reduction. Our idea comes from the similarity between a sparse feature matrix extracted from multivariate time-series and a term-document matrix. We conducted experiments with synthetic and simulated data, and confirmed that the proposed method successfully detected anomalies in multivariate time-series signals. Especially, it demonstrated superior performance with anomalies in which only relationships of time-series patterns are changed from reference data (multivariate anomalies).
Keywords :
computerised instrumentation; data mining; fault diagnosis; matrix algebra; sensors; time series; data mining application; feature extraction; generative signal models; multivariate time-series; semi-supervised anomaly detection; sensor data; sparse feature matrix; sparse representation; term-document matrix; time-domain correlations; Correlation; Feature extraction; Principal component analysis; Semantics; Sparse matrices; Time-domain analysis; Vectors;
Conference_Titel :
Systems, Man and Cybernetics (SMC), 2014 IEEE International Conference on
Conference_Location :
San Diego, CA
DOI :
10.1109/SMC.2014.6974327