An entity is a piece of information to be extracted from a document. Entities can have properties like required or maximum unique occurrences.
Annotation, label, entity value
A specific instance of an entity like an annotation or a prediction. E.g.
2022-02-13 is a value of entity
The process of marking or highlighting values of entities in a specific document.
In Metamaze, a document is a logical set of pages that is assigned exactly one document type. E-mails are treated as documents as well.
An upload is a logical set of one or more files to be treated together.
Entity consisting of text that is recognised on the document.
Any entity to be recognised that is not text. For example: signatures, handwriting, stamps, logo’s, …
A group of entities that belong together. For example: address consisting of street, number, city, … or an order line consisting of item description, item amount, item price, total price, …
The process of merging and splitting individual pages and files into one or more document(s).
Standardising raw input values into a structured format. For example standardising a date like “April 24th, 2020” to “2020-04-24”. Applies to dates, numeric values, currencies, …
Optical Character Recognition is the process of converting scanned images to actual text.