M-Files Artificial Intelligence now makes Metadata Suggestions

M-File’s next release will include some powerful artificial intelligence services to help end users with metadata suggestions.

I mentioned in another article that the power behind IML’s ‘Intelligence Components’ comes from the integration of Abbyy’s Artificial Intelligence (AI) technologies into M-Files. Based on textual content capture and OCR these AI technologies interpret text using algorithms that analyse the meaning of the words and the relationships between them allowing real time classification of complex and unstructured data.


Even though it sounds complex, it couldn’t be simpler for the end user. Drag a document into the vault and you’re presented with ‘tags’ or ‘suggestions’ that you can click on to populate the metadata fields. The ‘tags’ are based on information in the document that matches metadata already in the vault. For example, your vault may have a list of suppliers stored as metadata to use when adding invoices. When AI reads the invoice, because it matches the supplier name with the one stored as metadata in your vault, a tag is created as a suggestion for you. All you need to do is click on it to add it as metadata, much faster, easier and more accurate than clicking in the field and typing the name to select it.

It doesn’t stop there either; AI based metadata tags can be invoked on existing documents when a property is changed. This is especially useful for business processes where metadata is added or changed along the way in a workflow. Going back to our invoice example, the initial metadata suggested when added didn’t include a description as this is to be populated by the person approving it.

If you want to know more about how to setup M-Files Intelligence Service to make suggestions when adding documents or when a property is changed, please contact us.