Rule set definition overview

While data rules capture an understanding of data quality at the columnar level (as they are executed or associated with one or multiple columns) and provide information about which records meet or do not meet an individual rule, they do not capture how a record within a data source conforms to multiple data rules, for example, how many rules does a specific record break, nor do they identify the overall quality or confidence in a data source. Rule sets provide the capability to achieve this broader, more holistic, view of a data source and its records by executing and evaluating multiple rules together against individual records. The output from rule sets provides a view into your data at several levels:

Rule level

As with individual Data Rules, this view provides insight into the exceptions to each rule within the rule set. For example, "record 1 did not meet rule A."

Record level

Looking across the entire record, this view provides insight into the cases where a record had an exception to one or multiple rules. For example, "record 1 did not meet rules A, B, and C."

Source level

Looking through the entire data source, this view provides summarized insight into how many records have one or multiple rule exceptions, what the mean number of record exceptions are, what the standard deviation is from the mean, and, when compared to a baseline run, whether the source has improved or degraded in quality.