What is a doc?

It sounds like a rhetorical question coming from an app built around documents but there are a few key things to understand about working them in Docugami.


Docs refer to documents, specifically, written documents that are used in business and other institutions. 

Written business documents tend to be unstructured, in wildly varying forms. An important document type may take different visual layouts, like brochures, legal briefs, or research reports, yet exist for the same purpose and hold similar content.

Docugami currently works on .DOC. DOCX. and PDF files and can detect and decipher most written English material including formatting such as lists and tables. It cannot process images, handwriting or non-English language or non-standard formatting nor can it use encrypted documents.

At the core of our understanding of documents is DOCUMENT TYPES. Common types of documents might include RFPs (Request for Proposals), supplier agreements, customer agreements, employment contracts, licenses, service agreements, statements of work, and leases. Each document within any single document type may have different styles and structures, but they likely have creative patterns for the information, organized uniquely by the author.

By organizing documents by type, Docugami's AI can interpret parallel types of data in these similar docs which is why documents are organized in Docugami as Doc Sets. So when you upload documents it is helpful to organize similar doc types into their own folders and to be sure there are at least six documents of any given type so that we have a large enough sample to learn from.