unsupervised multivariate TS anomaly detection
andreaceruti opened this issue · comments
Hey, sorry if this is a stupid question but I am new to matrix profile algorithms.
My task is to find anomalies from unsupervised multivariate (10 dimensions) time series data coming from industrial sensors. I have read your documentation but I did not find any tutorial about this and so I am having some difficulties.
Is it possible to have some references about anomaly detection on multivariate and unsupervised time series data, so that I can try your algorithms? I am currently using autoencoders and they works fine, but I was wondering if matrix profile could overcome the results I have