The TD_IQR function creates a two-layered ART table. The primary result set contains the list of anomaly data points. The secondary result set contains the number of anomalies. The primary result set is accessed using a SELECT statement. The secondary is accessed using the TD_EXTRACT_RESULTS utility function.
RETURNS TABLE Schema for Primary Layer| Name |
Data Type |
Description |
|---|
| derived_series_identifier |
VARCHAR |
The resultant series identifier inherited from the SERIES_ID. |
| ROW_I |
Varies |
Row identifier of the ordered result sets. |
| Payload_n |
Varies |
Input series passed back with the result. |
| Payload_n_Anomaly |
Integer |
Each data point is checked to determine if it is an anomaly. A value of 1 means it is not an anomaly. A value of -1 means it is an anomaly. |
RETURNS TABLE Schema for Secondary Layer| Name |
Data Type |
Description |
|---|
| Payload_n_NUM_OUTLIERS |
Integer |
The number of anomalies in input data, and those that do not belong any clusters. |