What is a Data Cluster?

Sometimes you need to extract more than one set of fields from a set of documents. This is usually when not all fields are related to each other, rather there are sets of fields which are meaningfully linked to each other. So you can use one Data Cluster to represent each set of related fields you want to extract. For eg: if you want to extract the contents of an invoice, you might define the invoice heading details like invoice number, date, customer name, etc as one data cluster and the line items of the invoice like item, qty and amount as another data cluster.

Use data clusters to simplify your extraction work and lower the effort you need to label all the data you need.