A Privitar Schema is a description of the tables and columns of an input dataset. It defines the tables within the dataset, the names of the columns within each table, and the data types (for example, text, date, integer) for each column. It may also hold other information that is relevant for data processing purposes. For example, the date format or whether a column is optional or not.

Schemas can be reused, which is convenient when de-identifying the same dataset for multiple requests and use cases. They can be created manually via Privitar’s UI, imported from external sources such as Hive, inferred from files, or created programmatically via Automation REST APIs.

Return to glossary