SharePoint Syntex is a service provided by Microsoft to assist organizations. By design, it is a tool to help companies manage a large amount of content efficiency and Automated content processing. It uses advanced Artificial Intelligence (AI) and Machine learning (ML) to harness an organization’s expertise and content into knowledge. 

This allows an organization to find and manage its business contents easily and convert that into knowledge at scale. It allows them to streamline everyday business processes and tasks while reducing compliance and security risks by applying sensitivity and retention labels automatically.

Types of AI models in SharePoint Syntex

Currently, Microsoft provides two types of AI models in SharePoint Syntex.

1. Document The Understanding Model.

This type of model is used when you want to extract information from unstructured documents such as Contract Agreements, SLA (Service Level Agreements), SOW (Statements of Work), letters, etc., where the text entities you want to extract reside in sentences or a specific section of the the the document.

For example, a “Contract Agreement” can be written differently. Still, information exists consistently in the document such as Contract Agreement, effective as of followed by an actual agreement start date, This Contract Agreement start date & Contract Agreement end date is followed by an actual end date & start date of the agreement. It is made by and between followed by the agreement party’s name.

These models are created and managed in a SharePoint content center site.

2. Form Processing Model.

This model is used when the information is documented in a structured or semi-structured manner that follows a pre-arranged format like a tax Invoice and custom client build forms. Generally, in a tax Invoice, the entities such as client address, tax Invoice number, etc. are in the same location on the Invoice.

These models are created in PowerApps AI Builder, but the creation is initiated directly from a SharePoint document library.

The Business Use Case Of SharePoint Syntex

Irrespective of the industry, documents such as Contracts, SLA, Invoices, Forms, etc., are produced that contain business-critical information.

Handling hundreds of documents between clients and businesses, there is the need to uniquely categorize them while extracting key information such as the client’s name, client’s email, Contract agreement due date & end dates, etc., to create different types of automation process such as due date notifications as defined in the contract agreement and apply sensitivity and retention labels automatically.

I’ll walk you through a business use case that needs to identify the Contract Agreement and show you how you can achieve the business requirement with SharePoint Syntex.

NOTE: I’ll use the document understanding model type to create and train the AI model and extract the required information from the Contract Agreement documents.

Organizational benefits of using SharePoint Syntex:

Scenario #1

Tracking Information From Contracts With Document Understanding Models

 Observe a contracts management scenario, in which a process can be set up to identify your company’s contracts with other companies or individuals. A model can be set up to extract key information from those contracts, such as client name, costs, dates, and length, and is then added to the library as fields for a quick view. To ensure appropriate organization compliance, add a retention label on the document to ensure documents cannot be deleted before a specific length or period. 

What benefit do I get? 

• Save time and money by automatically extracting data from the contracts instead of doing it manually. 
• Ensure better compliance by using retention labels to ensure that the contracts are retained appropriately.

Scenario #2

Automate Order Processing To Reduce Manual Processing Of Customer Orders 

Observing a supply chain scenario where the Logistics Manager wishes to reduce errors caused by manual data entry and review of customer orders. Syntex allows you to upload orders from email, fax, or paper and applies AI and machine learning to validate the order information, extract core data and automatically push it into their ERP system for order fulfillment and reconciliation. 

What benefit do I get? 

• Order and shipment accuracy will increase. 
• Delays in invoicing or payments will decrease. 
• Fees or penalties related to shipment errors will be reduced.

Scenario #3

Avoiding Risk With Document Management, Records Management & Compliance Processes 

SharePoint Syntex allows you to set up compliance processes to capture, classify, audit, and flag documents and forms that need better governance. This mitigates end users manually tagging and archiving content as Syntex auto-classifies content to simplify search experience, manage data volumes, and apply records management and compliance whilst promoting best practices. 

What benefit do I get?

• Projects have all the documentation required to ensure company compliance policies. 
• Employees can easily discover the right information in the right context.
•  Compliance is upheld, and risk is reduced.

How can we do this?

Step 1 – Subscribe and activate SharePoint Syntex. 

• Log into your Microsoft 365 organization account through this.  

https://www.microsoft.com/en-us/microsoft-365/enterprise/sharepoint-syntex?activetab=pivot:overviewtab

• Once you sign up for a trial version or into a valid subscription, activate the Microsoft 365 admin center feature.
• Go to settings –> Organizational knowledge and click on the “Automate content understanding.” 
• Once you see the welcome screen, click on the get started button and select the configure form processing

Step 2 – Create a Content Center site

• After successfully activating SharePoint Syntex, create the Content Center SharePoint site to house the model

NOTE: You can create multiple Content centers based on your organizational requirement.

• Once the site has been created, navigate to the brand-new SharePoint Content Center site. 

Step 3 – Create content types (optional)

NOTE: – This step is optional as you can create a new content type while creating your model on the Content Center, so feel free to skip this step and move to step 4.

Step 4 – Create Models and train them.

• Task 1 – Create models

Now I’ll create the Contract model called “Contract Agreement” and associate it with the content type I did not create in the previous step (step 3). If you skip the step 3, select the “Create a new content type” under the Associated content type

• Navigate to the Models tab from the top navigation bar on the Content Center SharePoint site.
• Click on the + Create a model to create the Test Sharepoint syntex model and select the “Create a new content type” option. If you created a content type in the previous step, select the created content type under the existing content type.
• Click Create

You will be able to see a newly created model in the document library on the Content Center with the extension called “Classifier extension”, and “Test Sharepoint Syntex Classifier.”

Repeat the same process to create the content type (how many contents you have).

• Task 2 – Add example files

In this task, I’ll upload sample contract documents into the Training file library to train the model, as shown in the following figures.

The minimum requirement is to upload at least five positives and one negative sample. File types you can upload include PDF, JPEG, PNG, etc.

• Task 3 – Classify the documents and run training

I’ll now add intelligence to the contract content type by going through the documents and labeling them with an explanation.

Next, I’ll go through each of the six files and mark each file as an example of the contract or not by selecting Yes or No. 

Once you complete this task, you will see your results as follows.

• Task 4 – Provide an explanation

After identifying the documents correctly, I’ll start to train the model with my explanation as to why I have labeled some documents as Positive and others as negative. Explanations help the model distinguish the contract Agreements from other types of documents.

Microsoft has provided two options to create explanations.

1. Blank – In this option, we can create explanations using a blank template as a user.
2. From a template – in this option, Microsoft has provided a couple of predefined templates, such as date, currency, number, SSN, credit card, etc., to create the explanation.

I’ll use the blank option to provide my explanation as follows.

After adding my explanations, I can train these files by selecting Save and Train Model.

NOTE: The more types of explanation you add will produce greater accuracy when the machine classifies the documents.

In the figure below, you will see the document classification accuracy is 100; the document evaluation is matched based on the two explanations.

In addition to the above explanation, I have added another explanation called “Contains other Key Words.” In this explanation, I have added keywords that must exist in the documents (e.g., date, project name & Ref number), so when the machine classifies the documents, it scans through each document to see if these defined keywords exist in them.

Next, I’ll test the classification by uploading a contract document and other documents to see whether my model is smart enough to classify the document correctly.

Yes! My classification works as expected. Note how it has classified the TTQ-2313R1 PORT VIEW Mosque, Journeywork. Pdf as positive and the other document as negative.

• Task 5 – Train the model to extract the required metadata

Now let us extract the Reference Number, date, and project name from the documents. I’ll explain later why I have extracted these bits of information from the SLAs.

Click on Create Extractor -> Ref and give the new entity extractor a name. 

I’ll go through the positive sample files and select the value in the document I want to populate in the content type column. If the sample file is a negative sample, then select tick on the “No label”, otherwise, for positive samples (which is true in this case, TTQ-2328Add1/E1028/NK, is the Ref) click the Save button.

I’ll then train my model by adding an explanation as to why I have decided why TTQ-2328Add1/E1028/NK  is the reference number in the above document. The reference number in this document is followed by the keyword is ‘Before Label Ref:’

Once you save the explanation, you will see the evaluation and the accuracy of the explanation as follows.

If your explanation is identified correctly, the label you have added marked in Blue and the label added by classification marked in Green are matched together.

I have added similar extractions for the Date and Project Name. The list of extractions and the accuracy are shown in the figure below.

• Task 6 – Apply the model to libraries

In previous tasks, we have added intelligence to our documents, and now it’s time to apply the model to a SharePoint document library to see how it works in the SharePoint library.

First, I’ll apply the trained model to a document library, as shown below and upload some new documents and see how the model behaves.

Select the site and select the document library after clicking ‘Add’. Now it is added to the library.

Click Go to the library.

Now move to document library, upload document after some time AI Classify and extract details mentioned below.