Microsoft Azure has Cognitive Services available that can be used for auto-classification of files and images. This include the following features:

• Natural language processing skills include entity recognition, language detection, key phrase extraction, text manipulation, sentiment detection, and PII detection.

• Image processing skills include Optical Character Recognition (OCR) and identification of visual features, such as facial detection, image interpretation, image recognition (famous people and landmarks) or attributes like image orientation.

Here is an example of language detection, key phrase extraction, and sentiment detection.

Entity extraction automatically identify the tags or metadata that best describe the content. This could for the above example be:

• Contoso Steakhouse

• NYC

• John Doe

• Sirloin Steak

• URL

• Phone number

• Email

If you have Azure SQL databases or other systems that contain additional data about the extracted entities, then these can be auto-queried to add additional data. This could for Contoso Steakhouse be the address, geolocation, ratings, reviews, etc, For NYC, it could be State, Country, and Continent. Similar for the others.

To leverage Azure AI for text analytics, you need to send your information to Azure. This could be an app built for a specific purpose, or files stored in specific M365 libraries or folders are sent to Azure for processing before being returned to the storage location with auto-extracted metadata.

The above approach provides the following benefits when combined with other M365 features:

For users:

• Microsoft Viva Topics to discover relevant information and knowledge gathered by AI

• Search with metadata filters to better find information

For work groups:

• Microsoft Teams/site templates with automated governance and default metadata that files automatically inherent

• Azure AI and/or SharePoint Syntex to automate metadata extraction for key libraries and folders

• Trainable classifiers to automate the identification and classification of records and/or sensitive information for key libraries and folders

For the organization:

• Automated information governance and protection since the auto-extracted metadata can be used to auto-apply a the appropriate Record label and/or Sensitivity label.

• Improved knowledge sharing due to search being enriched with metadata filters for basic search, and Viva Topics help users discover relevant information and insights.

Azure AI can also be used for fast-and-smart migrations from legacy systems to M365 or Azure – check out this blog post that describes this in more detail.

Please feel free to contact us if you need help leveraging Azure AI for M365 Information Governance and Protection.