Data stream anomaly detection through principal subspace tracking

Pedro Henriques dos Santos Teixeira, Ruy Luiz Milidiú
2010 Proceedings of the 2010 ACM Symposium on Applied Computing - SAC '10  
We consider the problem of anomaly detection in multiple co-evolving data streams. In this paper, we introduce FRAHST (Fast Rank-Adaptive row-Householder Subspace Tracking). It automatically learns the principal subspace from N numerical data streams and an anomaly is indicated by a change in the number of latent variables. Our technique provides state-of-the-art estimates for the subspace basis and has a true dominant complexity of only 5N r operations while satisfying all desirable streaming
more » ... onstraints. FRAHST successfully detects subtle anomalous patterns and when compared against four other anomaly detection techniques, it is the only with a consistent F1 ≥ 80% in the Abilene datasets as well as in the ISP datasets introduced in this work.
doi:10.1145/1774088.1774434 dblp:conf/sac/TeixeiraM10 fatcat:m6oui6ezojhvxihyoorkd7gq6a