Organise your SharePoint data with Syntex
What is Syntex?
Microsoft Syntex is a content understanding, processing, and compliance service that uses intelligent document processing, content artificial intelligence (AI), and advanced machine learning to automatically and thoughtfully find, organise, and classify documents in your SharePoint libraries.
How does this help?
Rather than clicking and sorting through hundreds of thousands of files, Syntex extracts, analyses, and categorises the data for you.
With these capabilities, you can dive deeper into your content to truly understand it, and you can turn the information into meaningful insights that your organisation can use to make informed business decisions.
How do you use it?
Understanding your content with Syntex begins with models. Models let you identify and classify documents that are uploaded to your SharePoint document libraries, and then extract the information you need from each file. The type of model you choose will depend on the file type, the format and structure of the files, the information that you need to be extracted, and lastly where you want to apply the model.
What custom models are available with Microsoft Syntex?
There are 3 different types of custom models you can use to understand the layout of your files from example documents. The models learn to look for the data you need to extract from similar documents.
Namely the 3 model types are: Unstructured document, Freeform document, Structured document processing.
1. Unstructured document processing
Works best with letters and contracts
This method is used to automatically classify files and extract information. The documents must have text that can be identified based on phrases or patterns.
These created models are managed in the content centre in SharePoint online. You can add classifiers and extractors to your unstructured document processing models to further utilise Syntex's capabilities:
- Classifiers: used to identify and classify documents in the document library. For example, a classifier can be "trained" to find all contract renewal documents that are in the library. The contract renewal content type is defined by you when creating the classifier.
- Extractors: pulls information from the documents. For example, each contract renewal document identified in the document library will have columns to show the Contract start date and client name on each document.
Microsoft states 5 example files should be used to train and test your classifiers and extractors in your model. Example files provide your model examples of what to look for when trying to identify and extract data from the files. For example, you would train your contract renewal classifiers and extractors with examples of contract renewal documents your company works with. You can also use example files to test the effectiveness of your model.
You may also use unstructured document processing to apply retention labels. On the model's homepage you can apply pre-exisiting retention labels along with collecting data to prevent deletion of the files for the retention period.
2. Freeform document processing
Works best with letters and contracts
This enables you to automatically extract information from unstructured and freeform documents.
You can use freeform document processing to create processing models that use machine learning technology to identify and extract key-value pairs and table data. The AI automatically extracts the text and key-value pairs and/or tables from various sources such as mail fax and emails that are uploaded to SharePoint, to help data processing for your company.
3. Structured document processing
Works best with forms and invoices
Structured document processing, similarly to Freeform processing, uses machine learning technology to identify and extract key-value pairs and table data from structured or semi-structured documents, such as forms and invoices.
You may also use structured document processing to apply retention labels. On the model's homepage, you can apply pre-exisiting retention labels to prevent deletion for a defined period for example.
As both Freeform and structured document processes are similar, view the table here that details the differences to help understand which best suits your needs.
How do you get started?
Firstly, admins need to decide which SharePoint Site will need the Syntex features and "enable document processing". Each user that needs to use Syntex must have Syntex license assigned, and for Structured and Freeform document processing, AI credit will be required.
Once you have enabled document processing and licenses been assigned, admins can go to the Microsoft 365 admin centre then open the setup page and in the "view files and content section", follow through the automate content understanding setup guide.
Please note, read-only or Sealed content types cannot be updated or used in a model.
What do you need?
- On the SharePoint site(s) Enable document processing must be turned on
- Syntex license for the user or users
- AI Credit if structured or freeform document processing will be used
Next steps
Need support? As a longstanding Microsoft Partner with an Infrastructure (Azure) Designation and an AVD Advanced Specialisation, Advanced are best placed to help your business make the most out of your technology investments and achieve your objectives. Simply get in touch today if you’d like advice or help getting started.
Written by Daniel Bereczki, Cloud Solutions Architect, Advanced