Build AI models embedded in the applications to the needs of your business

Disclaimer: This is my personal blog, therefore anything I post, share, and comment don’t reflect the view of my employer.

Contributors:

• Kyle Akepanidtaworn, Cloud Solution Architect (Data & AI)
• Boonsita Vatcharakomonphan, Account Technology Strategist Intern
• Napat Ratanakul, Account Technology Strategist Intern
• Noppadol Rattanawisadrat, Program Trainer in Microsoft Education

During Microsoft Inspire 2019 Corenote with Satya Nadella, the CEO discussed Microsoft’s mission, solution areas and the impact our employees and partners can have on the 7 billion people on the planet. There was a conversation between Ryan and Eric with how G&J Pepsi leverages AI in the context of PowerApps, and that feature is called “AI Builder” in PowerApps.

For those of you who are not yet familiar with PowerApps, “PowerApps is a suite of apps, services, connectors, and a data platform that provides you with an opportunity to build custom apps for your business needs. By using PowerApps, you can quickly build custom business apps that connect to your business data that is stored either in the underlying data platform (Common Data Service) or in various online and on-premises data sources (SharePoint, Excel, Office 365, Dynamics 365, SQL Server, and so on). Apps that are built by using PowerApps provide rich business logic and workflow capabilities to transform your manual business processes to digital, automated processes. PowerApps simplifies the custom business app building experience by enabling users to build feature-rich apps without writing code. PowerApps also provides an extensible platform that lets pro developers programmatically interact with data and metadata, apply business logic, create custom connectors, and integrate with external data.” You can see more details in the following links:

What is an AI Builder?

As per the focus of the article on “AI Builder”, I will stop the narrative introduction of the PowerApps right there. “AI Builder is a new Power Platform capability that allows you to easily automate processes and predict outcomes to help improve business performance. AI Builder is a turnkey solution that brings the power of AI through a point-and-click experience. Using AI Builder, you can add intelligence to your apps even if you have no coding or data science skills.”

Remark

• AI Builder is a preview feature.
• Preview features aren’t meant for production use and may have restricted functionality. These features are available before an official release so that customers can get early access and provide feedback.
• AI Builder is enabled by default in your environment if your environment is eligible for the feature. Administrators can control the feature availability for their environment using the Power Platform Admin center. More information: Enable or disable AI Builder feature

In AI Builder, you can create tailored AI models to automate processes and find insights on 4 main tasks:

1. Binary Classification — Classify the elements of a given set into two groups (predicting which group each one belongs to) on the basis of a classification rule.
2. Form Processing — Capture information entered into data fields and convert it into an electronic format. This can be done manually or automatically, but the general process is that hard copy data is filled out by humans and then “captured” from their respective fields and entered into a database or other electronic format.
3. Object Detection —Detect instances of semantic objects of a certain class (such as humans, buildings, or cars) in digital images and videos. Well-researched domains of object detection include face detection and pedestrian detection. Object detection has applications in many areas of computer vision, including image retrieval and video surveillance.
4. Text Classification — Assign a text corpus to one or more classes or categories. This may be done “manually” (or “intellectually”) or algorithmically.

Now, let’s get back to Microsoft Power Platform Admin Center. Choose {Your Environment} → Setting → Under “Products”, choose “Features” → Enable AI Builder.

Okay, if you’re at this step, you’re all good. Thanks the contribution from Napat Ratanakul for debugging!

Here are important sites:

• web.powerapps.com — On this site, you can open apps, specify the type of app that you want to create, share your app, and create data connections and flows. To use this site, you’ll need to log in by using your organizational account.
• PowerApps Studio — On this site, you build apps by configuring user interface (UI) elements and Excel-like formulas.

Get on web.powerapps.com & choose the environment you just created.

The next thing is to import the data: Data → Entities → Get Data. See the instruction here:

Esure that you successfully upload the data, and it’s showing in one of the “Entities”

Let’s now go to “AI Builder (preview)”, you should see the following screen.

As an example, we are going to work with “Text Classification.” Here’s what you may expect:

Provide “Model name”. It can be anything, really. However, the required data are text corpus & tags (view them as labels).

Currently, the text classification model supports 6 languages. Too bad for the Thai language…If your language is not applicable, you may want to consider other solutions or train your own model on Azure.

Wait for the model to train! Just several minutes. You should see the result below:

You can perform the quick test: let say I use the following article as a test case.

Of course, there are ways to improve the model, but our task is already completed! We, as a team, demonstrate an end-to-end text classification task using AI Builder. You can publish this model as an application or leverage the model through Microsoft Flow.