QUESTION: 88 Exam Topic: Implement AI predictions in Customer Insights

Your company uses Microsoft Dynamics 365 Customer Insights' out-of-the-box (OOTB) machine learning (ML) capabilities across product recommendation, customer lifetime value, and churn. Now the company wants to develop a new custom ML scenario that is not covered within the OOTB capabilities. The company does not have data scientists or ML practitioners. There is no time to train someone to develop a new ML scenario.

You need to recommend how to implement workflows that consume ML models.
Which two recommendations should you make? Each correct answer presents a complete solution.
NOTE: Each correct selection is worth one point.

You must register datasets in the workspace. The datasets can be represented in multiple formats including JSON and Parquet.
You can use Microsoft Azure Machine Learning designer to configure a batch pipeline that can be integrated into Dynamics 365 Customer Insights.
You can use a training pipeline to create or update an inference pipeline. Only batch inference pipelines are currently supported.
You can use Microsoft Azure Machine Learning SDK to develop a custom ML model, to consume in Dynamics 365 Customer Insights, and when using a batch inference pipeline.

Answer(s): B,D

Explanation:

B: Work with Azure Machine Learning designer
Azure Machine Learning designer provides a visual canvas where you can drag and drop datasets and modules. A batch pipeline created from the designer can be integrated into Customer Insights if they are configured accordingly.

D: Work with Azure Machine Learning designer
Azure Machine Learning designer provides a visual canvas where you can drag and drop datasets and modules. A batch pipeline created from the designer can be integrated into Customer Insights if they are configured accordingly.

Incorrect:
Not A; Batch pipeline requirements to integrate with Customer Insights
Dataset configuration
Create datasets to use entity data from Customer Insights for your batch inference pipeline. Register these datasets in the workspace. Currently, we only support tabular datasets in .csv format.

Not C: Batch inference pipeline
In the designer, use a training pipeline to create or update an inference pipeline. Currently, only batch inference pipelines are supported.

Using the SDK, publish the pipeline to an endpoint. Currently, Customer Insights integrates with the default pipeline in a batch pipeline endpoint in the Machine Learning workspace.

Reference:

https://learn.microsoft.com/en-us/dynamics365/customer-insights/azure-machine-learning-experiments

Reveal Solution Next Question

QUESTION: 89 Exam Topic: Implement AI predictions in Customer Insights

Your team developed and integrated a batch inferencing pipeline to Microsoft Dynamics 365 Customer Insights. This allows for a new custom machine learning (ML) model to run and deliver new insights to the unified profiles' data.

You need to make sure the workflow runs automatically with every scheduled refresh.

What should the team do?

The team needs to configure a Microsoft Power Automate flow to trigger a refresh.
The team can set up scheduled refreshes that will also run the workflows automatically.
The team needs to manually trigger the custom ML workflows.
The team can configure scheduled refreshes for Dynamics 365 Customer Insights from Azure Machine Learning where you developed the custom ML model.

Answer(s): B

Explanation:

Custom models lets you manage workflows based on Azure Machine Learning models. Workflows help you choose the data you want to generate insights from and map the results to your unified customer data.

Add a new workflow (see 11 below)
1. Go to Intelligence > Custom models and select New workflow.

2. Provide a recognizable Name.

3. Select the organization that contains the web service in Tenant that contains your web service.

4. If your Azure Machine Learning subscription is in a different tenant than Customer Insights, select Sign in with your credentials for the selected organization.

5. Select the Workspaces associated with your web service.

6. Choose the Azure Machine Learning pipeline in the Web service that contains your model dropdown. Then, select Next. Learn more about publishing a pipeline in Azure Machine Learning using the designer or SDK.

7. For each Web service input, select the matching Entity from Customer Insights. Then, select Next.

8. For Model Output Parameters, set the following properties:

Entity name for the pipeline output results
Output data store parameter name of your batch pipeline
Output Path parameter name of your batch pipeline
Model Output Parameter Pane.

9. Select the matching attribute from Customer ID in results that identifies customers and select Save.

10. Select Done. The Custom Models page displays.

11. Select the vertical ellipsis for the workflow and select Run. Your workflow also runs automatically with every scheduled refresh.

Reference:

https://learn.microsoft.com/en-us/dynamics365/customer-insights/custom-models

Reveal Solution Next Question

QUESTION: 90 Exam Topic: Implement AI predictions in Customer Insights

You are planning to use the AI-driven suggested segments, based on the customer activity data that is ingested to Microsoft Dynamics 365 Customer Insights.

You need to determine how the machine learning model analyzes the activity data to suggest segments.

Which three dimensions are used? Each correct answer presents a complete solution.

NOTE: Each correct selection is worth one point.

Customer ID
Monetary Value
Recency
Activity Type
Frequency
Unified Activity entity

Answer(s): B,C,E

Explanation:

Suggested segments based on activity
Discover interesting segments of your customers based on customer activity data that is ingested to Customer Insights. Examples of activity data are transactions, support call duration, purchases, or returns. To suggest segments, activity data gets analyzed for recency, frequency, and monetary value (or duration).

Required data
Suggestions are generated based on the selected input data.
* Dimensions to include: Recency, frequency, or monetary dimension, depending on your business requirements.
* Etc.

Reference:

https://learn.microsoft.com/en-us/dynamics365/customer-insights/suggested-segments-activity

Reveal Solution Next Question

QUESTION: 91 Exam Topic: Implement AI predictions in Customer Insights

You created several segments in Microsoft Dynamics 365 Customer Insights.
You need to perform a segment overlap analysis to understand what these segments have in common.

Which two actions can you take? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.

You can save the overlapping members as a new segment
You can export the analysis output for further review outside Dynamics 365 Customer Insights.
You can analyze how specific values affect the overlap results.
You can view the analysis output to determine the count of members included in only one segment.

Answer(s): C,D

Explanation:

Segment overlap
Segment overlap analysis shows how many and which customers are common to two or more segments. For example, how a segment of frequent customers overlaps with a segment that contains customers that are satisfied with your service or product. You can also analyze how the overlap changes for specific attributes. (C)

View and optimize an overlap analysis
1. After completing the analysis, find details on this insight on Segments > Insights (preview).

2. The number of members overlapping the segments selected for analysis.
(D) The number of members included in one of the segments but not in the rest of the segments.
-If you selected fields while configuring the overlap analysis, you'll find them in the corresponding tabs. You can use the filter dropdown to select any attribute level of interest and the table at the bottom will show the corresponding data.

Reference:

https://learn.microsoft.com/en-us/dynamics365/customer-insights/segment-insights

Reveal Solution Next Question

Free MB-260 Exam Braindumps (page: 13)

QUESTION: 88 Exam Topic: Implement AI predictions in Customer Insights

Explanation:

Reference:

QUESTION: 89 Exam Topic: Implement AI predictions in Customer Insights

Explanation:

Reference:

QUESTION: 90 Exam Topic: Implement AI predictions in Customer Insights

Explanation:

Reference:

QUESTION: 91 Exam Topic: Implement AI predictions in Customer Insights

Explanation:

Reference: