Use regular expressions in Document Processing conditions
When setting up a template for Document Processing it would be good to have more control over indexing and identification of documents. Using regular expressions would be great for indexing fields where you will not always know what text you want to replace, you could find a pattern and then replace that. It also goes for the identification of a document. When you are looking for something on the document to identify what kind of document you could search for a regex pattern to identify the document for processing.
Scott Bradford commented
Was looking at doc processing solutions recently and found several good ones. The one listed here uses Fuzzy Regex, which should be pretty darn close to what you're describing. Hope this helps https://blog.bisok.com/general-technology/what-is-intelligent-document-processing