Seasonality in Billing Anomaly Detection

Oracle Revenue Management and Billing enables you to build a data model considering seasonality and accordingly predict the billing anomalies. Seasonality is a characteristic of a time series in which the data experiences regular and predictable changes that recur every calendar year. Any predictable fluctuation or pattern that recurs or repeats over a one-year period is said to be seasonal.

We offer this feature based on the assumption that an account's behavior would be similar for a given month across the years. To enable the seasonality feature in billing anomaly detection, you need to set the Seasonal Prediction Flag option type of the C1-BRMLINOPS feature configuration. It is used to indicate whether seasonality should be considered while extracting the data for model building. The valid values are Y and N. If you set this option type to Y, the system considers the following fields of the respective entities during model building:

If you do not specify the value for this option type, by default, it is set to N. If you change the value of this option type, you need to run the C1-MLDPR and C1-MLMCR batches once again to rebuild the machine learning training model.