Relationship Between Field ID and Table ID

• Published on Sep 17, 2024

• 1 minute(s) read

Overview

Documents can contain normal fields, table fields, or both. The Supervision tasks associated with a document are dependent on the types of fields it contains. In each of the following scenarios, the field type(s) in a document are pre-defined by the user in the respective Semi-structured layout.

For more information, see Creating Semi-structured Layouts.

Documents that only contain table fields

• Table fields are characterized by the fact that each field, or cell, has a relationship with a column header, and each series of related fields constitutes a row.

• Tables do not need to have a standard grid format: if a table has distinct columns, but the individual values in a row don’t share the same alignment, it still qualifies as a Table.

• The Table ID task is the only Supervision task required to identify data on this kind of document.

• Note that a Table ID task will be generated whenever a document contains a table, even if the machine is able to confidently identify the table within the document. This functionality allows you to verify the table’s structure before sending the document to Transcription.

For more information, see Table Identification.

Documents that only contain normal fields

• Normal fields are text, signatures, and checkbox fields – anything that is not a table field.

• The Field ID task is the only Supervision task required to identify data on these kinds of documents.

• If the machine is able to confidently identify all of the fields on a document, then Hyperscience will not generate a Field ID Supervision task.

For more information, see Field Identification.

Documents that contain both field types

• Field ID and Table ID are both necessary Supervision tasks in order to identify all of the data on these kinds of documents.

• First, the user will perform Field ID and then they perform Table ID.

• Due to the presence of a table in the document, a Table ID task will always be generated so that you can confirm what the machine has identified before passing the document along to Transcription.

For more information about both of these tasks, see the articles linked above.