Microsoft Power Automation and Power Builder

AI Projects can leverage PowerAutomation from Microsoft to determine the Title, Author, and Date of works submitted.

Power Automate (https://powerautomate.microsoft.com/) requires credits that can be requested from UC DTS.
AI Builder

1. AI Models The first step is to create a model. Microsoft provides various templates to start with model creation.

For our Model, I chose Document Processing, since we want to extract meta-data from the document.

2. Training Document and Identifying information to Extract The second step is to to define a training corpus. I chose 10 ETD pdfs since each PDF has a title page with a structured layout for Title, Author, and Date of works. For each document it is required that you draw a box around a field and label it.

Once you have trained your Model, you can do a quick test.

3. My Flows and Cloud Flows. The third step is to create a workflow that will use our model when triggered. I chose to Manually trigger a workflow. These workflows are useful in that they completely integrate with all of the Microsoft products.

This Flows can be RUN using our Model to scan the documents, extract the meta-data, and send an e-mail.

4. Flow Checker Feedback Flows I still need to do more research on Feedback Flows and how to help the model learn after each run.

Here is an article on how to use form processing to extract data from documents triggered manually.

How to : https://learn.microsoft.com/en-us/microsoft-365/community/machine-learning-and-managed-metadata