Power Apps – AI Builder – Part 1

In this blog, we will be looking at Power Apps AI Builder, which is a recent addition to Microsoft Power Platform and allows app makers and users to add intelligence to their apps by creating AI models tailored to their requirements.

Getting started is straightforward. Just head onto the power apps maker portal (https://make.powerapps.com), select the environment which you would like to work in, navigate to AI Builder section in the navigation bar and select “Models” menu.

Here we can start creating our own model. We can begin by clicking “Build a model” button. The next step is to select the type of model that we will be building. There are many types of models to choose from, each serving their own purpose and catering to a specific use-case. For demonstration purposes, we will be making a “Form Processing” model. Next, we will need to name our AI Builder model and click on “Create”.

We will then be greeted with the “Choose information to extract” page. On this page, we need to outline the key pieces of information which must be extracted from the documents. Click on “Add” to start adding items.

There are 4 types of fields which can be extracted from documents:

1. Field: Used for extracting printed or handwritten text. (For example: Name, Date, Age, Address)
2. Single page table: Used for extracting information present in tabular format (For example: list of employees, sales order items list)
3. Checkbox: Used for extracting selection-based information like checkboxes and radio buttons (For example: marital status, agreement)
4. Checkbox: Used for extracting selection-based information like checkboxes and radio buttons (For example: marital status, agreement)

Let us add a few Fields to our model. Here is how the list looks like after adding fields: We are good to go with our Fields. Now, let us add some tables to our model. Click on “add”, select “single page table”, click “next”.

To add tables, we first need to give a name to our table. Next, we need to define the column names for this table. To rename “Column 1”, click on the first column header and select “Rename column” option for the dropdown menu. Now we can update the name of the table’s default column. We can continue to add more columns to our table until we have extracted all pieces of information from the table in our documents.

This is how the page should look like after adding columns to our table: For demonstration purposes, we have added two tables to our model per the tables present in the documents which we will be using to train our model. This has been great progress so far in our quest to add intelligence to our Power Apps. We will continue this journey in the next blog of this series where we will be looking at adding document collections to our model and training the model on those documents. Thank you for following along!

Join us next time, as we continue our journey of learning canvas apps.Click here to learn more about Imperium's Power Apps Services. We hope this information was useful, and we look forward to sharing more insights into the Power Platform world.

---