Describe different types of Models
We shall now describe models including business card reader, detection model, form processing model, and prediction model.
Business Card Reader
- First, you must use AI Builder to take a photo or load an image from a local device
- Then get it scanned to find business card-related information.
- If any is detected, the business card reader component extracts it and exports it.
Object Detection Model
In order to build the object detection model, sign in to Power Apps, and then in the left pane. Now select the down arrow to expand AI Builder. Thereafter select the Object Detection AI model type and enter a name for your model, and then select Create. When you create an AI Builder object detection model is to define its domain including Common objects, Objects on retail shelves and Brand logo.
Steps in building Object Detection Model include –
- Firstly, select the model domain
- In the second step, provide object names
- Next, enter names in AI Builder
- Then, select names from a database
- Thereafter, upload images
- Lastly, tag images
Processing Model
- Firstly the processing models allows us to read and save information from standard documents like invoices or tax documents.
- Secondly, after automating this process, we can save valuable time by reviewing, extracting, organizing, and saving the data automatically by using Power Automate and Power Apps.
- Next, train your model and define the information to be extracted from the forms.
- Then get accurate results tailored to your specific content.
- After you train and publish the model, we can use it in a flow in Power Automate or in a canvas app in Power Apps.
Prediction Model
- Firstly, the AI Builder prediction models analyze patterns in your data
- Secondly, it helps to learn and associate historical patterns with outcomes.
- Then the prediction model learns patterns and detects them in new data to predict future outcomes.
- Also, we may use the prediction AI model to explore business questions answered as one of two available options (binary), multiple possible outcomes, or where an answer is a number.
Types of Prediction Model
- Binary prediction: It is used when the question asked has two possible answers – yes/no, true/false, and more.
- Multiple outcome prediction: It is used when the question can be answered from a list of more than 2 possible outcomes.
- Numerical prediction: It is used when the question is answered with a number.
