BI Endgame – When AI meets BI
Extending
Artificial
Intelligence
models
BI
Some of the largest cloud vendors in our industry are rapidly developing AI capabilities that can be
extended to BI to uncover new frontiers of insight. We integrate with these AI capabilities in our data
ingestion process (https://docs.microsoft.com/en-us/power-bi/service-dataflows-overview) which can
be configured following these links: https://aka.ms/enableaiworkload & https://aka.ms/azuremlpbi
Image
recognition
and
sentiment
analysis
without
coding
Users looking to gain insight from unstructured data can
now use the AI from Azure Services to get capabilities such
as image recognition and sentiment analysis. And the best
part is that no code is required so BI users to discover
hidden, actionable insights in their data and drive better
business outcomes with easy-to-use AI without having to
write custom code.
These capabilities are enabled through integration with
Azure Cognitive Services (https://azure.microsoft.com/en-
us/services/cognitive-services/) to classify images, detect language sentiment, etc. Once this is
configured, cognitive services are accessible through the AI Insights browser, which is an editor for
dataflows. From this editor, a user can rate the sentiment analysis of a customer’s review or even detect
the language the text is in.
Introduce
predictive
analytics
in
BI
Users looking to identify patterns in large, noisy structured data sets can create machine learning
models directly using Automated Machine Learning (https://studio.azureml.net/), which is a cloud
predictive analytics service that makes it possible to quickly create and deploy predictive models as
analytics solutions. Models built by data scientists can now be easily shared with business analysts. This
makes collaboration among business analysts and data scientists easier and faster than ever before.
The AI Insights browser displays all ML models that have
been shared. These models can be introduced to
preexisting queries and will return the appropriate score
that the ML model calculates. Users can access existing
ML models via AI Insights in the editor or they can
navigate to any dataflow and build a new ML model
through AutoML: https://powerbi.microsoft.com/en-
us/blog/creating-machine-learning-models-in-power-bi/
As an example, a business analyst could leverage the automated machine learning technology to quickly
and easily build a model to predict how likely an open sales opportunity is to be won. This could help a
sales manager prioritize which high value opportunities to focus on and how likely they are to meet their
target.
Artificial
Intelligence
within
BI
New insights can also be generated on top of the data that is published to the cloud service. The
following are features that anyone accessing a BI report can take advantage of from their web browser.