ML-based rules

In this section you will find some guidelines to configure your model better and improve its forecasting accuracy.
You can for instance:

  • Monitor data assets on an hourly level if you need a more precise model when you receive hourly new data.
  • Choose the right sensitivity for your model, depending on the importance of your data and the jumpiness of its data or metadata. If you want to monitor only significant events and not smaller movements, you might want to lower the sensitivity to avoid receiving false-positive alerts. On the contrary, if any slight change in your data is critical, you might want to increase the sensitivity to capture it.
  • Improve the accuracy by providing feedbacks on alerts


Query optimization

In order to optimize the volume of data queried, Sifflet only queries by default the newly data ingested since the last rule execution.