Microsoft Ignite Virtual 2020: SharePoint Syntex

Announced at Microsoft Ignite Virtual 2020, is the exciting SharePoint Syntex. If you recall from last year’s announcements, Microsoft had excitedly introduced Project Cortex – an initiative with Microsoft 365 – with the vision of empowering people with knowledge and expertise in their apps by applying and leveraging advanced artificial intelligence (AI). SharePoint Syntex is the first product resulting from Project Cortex and will be available for all Microsoft 365 commercial customers to purchase at beginning of October 1, 2020.  

Most organizations have numerous databases and when applicable and/or specific content must be located, they turn to experts and information workers to locate this data. Large amounts of time is spent manually and methodically searching across these large databases. This is not only time consuming, but it is super inefficient. There is a theory that if an expert can teach students information processing skills, then that skill can be taught to a machine. With machine learning, experts can teach or train SharePoint Syntex to understand content like they do and what content they want captured. SharePoint Syntex can be taught to recognize key information and automatically tag content, creating a model based on the expert’s training. Throughout this process, the expert does not need to write or implement any code. Once this model is created, SharePoint Syntex will automate the capture, ingest, and categorize the content. From there, it will extract valuable information and transform it into metadata.

Why metadata? Because metadata is how content is managed, providing the capabilities for knowledge discovery, knowledge reuse, accelerate processes, and application of compliance policies and information protection by seamlessly integrating Microsoft Information Protection, Power Automate, and Microsoft Search.

The Nuts and Bolts

SharePoint Syntex Content Centre

The SharePoint Syntex Content Centre is a flow and combination of content management at large scale, metadata integration, workflow integration, and automated compliance delivery. The Content Centre takes what we do manually – read and process documents – and teaches it to the cloud. SharePoint Syntex then uses these insights, automatically recognizing content, extracting important information, and applying metadata tags. All this can be tracked and reviewed with the integrated visual analytics.

Content Processing

What types of content can be processed? SharePoint Syntex mainstreams advanced AI to process three types of content:

1. Structured or Semi-Structured Forms: these types of forms include invoices, purchase orders, and forms that might include common values like dates, figures, addresses, and names. Up until now, information was manually extracted from these forms then transferred to Excel or another database and then passed forward to the next responsible person or department.

This slow and tedious process is now performed with the powerful AI Builder that powers the form processing engine in SharePoint Syntex. This form processing engine is capable of automatically recognizing and extracting common values, as listed earlier, from structured and semi-structured documents.

These models are built without any code and provide reliable results, even with a small sample of documents;

2. Digital Images: SharePoint Syntex uses its visual dictionary, containing thousands of common and recognizable objects, to automatically tag images for search. Additionally, it can recognize handwritten text and convert this extracted data into tags for further processing; and

3. Unstructured Documents: organizations generate vast amounts of unstructured documents such as contracts, manuals, and resumes. Similar to machine teaching to build AI models with no codes, SharePoint Syntex, and be taught to read the content in the same way that you want it to be read. Metadata can be automatically created or suggested, customized Power Automate workflows can also be automatically suggested or created and compliance labels can be automatically attached for record management policies or to enforce retention policies. Automated security governance can take place with the addition of sensitivity labels to your models.

Microsoft Graph Content Connectors

When we think about content, we associate our content as being stored inside the Microsoft Cloud, which is true for the majority of an organization’s content, and with SharePoint Syntex, it is another content service storing an organization’s data in the Microsoft Cloud. However, content is stored and lives in repositories outside of the Microsoft Cloud which is why integrating remote sources is vital. Microsoft Graph Content Connectors will have additional capabilities such as file shares, Azure SQL, and third-party sources (like Box and IBM FileNet) integrated into Microsoft Graph, making them usable and searchable in Microsoft 365.

Taxonomy Services

Taxonomy services, also known as managed metadata services, provide the rules for the management of types of content, whether it is a document template or custom columns. With Taxonomy Services, your content tags are based on the definitions of the hierarchy of classifications and synonyms that you define.

SharePoint Syntex brings analytics to Taxonomy Services and delivers this report to the SharePoint Admin centre. This report provides the capabilities to monitor and provides analyzed information on term creation and usage throughout Microsoft 365.

Third-party taxonomies can be smoothly adapted into SharePoint Syntex by importing the new format for Simple Knowledge Organization Systems (SKOS) terms which provides the abilities to define richer import files, supporting the definition of custom properties and defining multiple terms sets while importing into a group.

Even more exciting, shared content types can be published from the central gallery through hub sites to SharePoint and Microsoft Teams. Sites connected to hubs will automatically receive updated and published content types. Streamlining, this now provides flexibility and rapid deployment across an organization.

Administration and Configuration

Making it easier for administrators of content services, SharePoint Syntex will be deployable on the first day by guiding administrators with wizards for configuration and deployment.

Managing content with SharePoint Syntex is simple due to automation, machine teaching, and advanced AI. Once manual and tedious, many of these processes are now being performed automatically with AI smarts, freeing up precious time to focus on other important issues.