Effortlessly auto-tag and organize SharePoint with AI and Flow

Table of Contents

1. Introduction
2. Setting Up the Library
3. Creating the Flow
4. Adding the OCR and Text Analytics Services
5. Joining the Key Phrases
6. Updating the Item
7. Testing the Flow
8. Troubleshooting
9. Conclusion

Introduction

In this article, we will explore how to auto-tag SharePoint content using AI. We will utilize cognitive services and flow to extract metadata from scanned images and add it as tags to the content. This process can greatly enhance the searchability and organization of documents in SharePoint.

Setting Up the Library

To begin, we need to set up a document library in SharePoint. We will be using the modern experience, but this process can also be done in the classic experience. We will add an additional multi-line column called "My Tags" to hold the generated metadata.

Creating the Flow

Next, we will Create a flow to automate the tagging process. We will start with a blank flow and use a SharePoint trigger to detect when a file is created in the document library.

Adding the OCR and Text Analytics Services

To extract metadata from the scanned images, we will use optical character recognition (OCR) through the computer vision API. We will also utilize the text analytics service to extract key phrases from the recognized text.

Joining the Key Phrases

The key phrases will be output as an array, but we want to work with a single piece of text. To achieve this, we will use the "join" operation to concatenate the key phrases with commas as delimiters.

Updating the Item

Once we have the key phrases, we will update the item's metadata in SharePoint. This will involve using the SharePoint "update item" action in our flow.

Testing the Flow

Before deploying the flow, it is important to test it to ensure it is working correctly. We will upload a document to the library and observe the flow in action. We will check if the generated key phrases are added as tags to the document.

Troubleshooting

If any issues arise during the testing or deployment of the flow, we will discuss common troubleshooting steps and potential solutions.

Conclusion

Auto-tagging SharePoint content using AI can significantly improve document management and search capabilities. By leveraging cognitive services and flow, we can streamline the process of extracting metadata from scanned images and adding it to the content. This enables more efficient organization and retrieval of information within SharePoint.

🌟 Highlights

• Auto-tag SharePoint content using AI
• Extract metadata from scanned images
• Add metadata as tags to content
• Enhance searchability and organization in SharePoint

FAQs

Q: Can this process be used with both the modern and classic experience in SharePoint? A: Yes, the auto-tagging process can be implemented in both the modern and classic experiences of SharePoint.

Q: What services are utilized to extract metadata from scanned images? A: Optical character recognition (OCR) and text analytics services are used to extract metadata from the scanned images.

Q: How are the key phrases added as tags to the content? A: The key phrases are joined into a single piece of text and then updated as metadata for the respective content item in SharePoint.

Resources:

• SharePoint
• Computer Vision API
• Text Analytics