Hacker News new | ask | show | jobs
by edublancas 1570 days ago
Congrats on the launch! As a former data scientist, I suffered from bad data on a daily basis. Can you provide some details on how anomalies are detected? Is it some kind of threshold-based approach defined by the user or are you running statistical analysis on user's data? Curious to learn more!
1 comments

Anomalies are detected based on a statistical analysis of the data and is measured in terms of standard deviations from the mean. We have lots of plans on improvements in the future and curious also to learn how you would approach this problem as well.