Lightweight Automated Feature Monitoring for Data Streams. (arXiv:2207.08640v2 [cs.LG] UPDATED)

Monitoring the behavior of automated real-time stream processing systems has
become one of the most relevant problems in real world applications. Such
systems have grown in complexity relying heavily on high dimensional input
data, and data hungry Machine Learning (ML) algorithms. We propose a flexible
system, Feature Monitoring (FM), that detects data drifts in such data sets,
with a small and constant memory footprint and a small computational cost in
streaming applications. The method is based on a multi-variate statistical …

arxiv data data streams feature lg monitoring