um_model: A function to retrieve the anomalies using rds.

In Dorialexander/UmbertoR: Detection of anomalous events in web logs data.

Description

A function to retrieve the anomalies using rds.

Usage

um_model(clean_log_data, knots = 3, min_observations = 7,
  min_anomaly_ratio = 5)

Arguments

clean_log_data	A dataset of logs aggregated by publications and per days already cleaned by um_clean
knots	The number of knots fitted to the knots. By default 3. A larger value may better fit log data with a long/complicated history, at the expense of being to overfitted to anomalous episode.
min_observations	The minimal number of observations to derive the model. By default 6 with 3 knots.
min_anomaly_ratio	The threshold of anomaly rate (by default 5 but can be any number between 0 and 100)

Examples

 um_model(clean_log_data, knots = 3, min_observations = 7, min_anomaly_ratio = 5)

Dorialexander/UmbertoR documentation built on Feb. 13, 2020, 12:41 a.m.