AI in Power BI
In part 1 and 2 of our Artificial Intelligence (AI) in Power BI series, we covered AI visuals and the use of Azure Cognitive Services to augment analysis with advanced data science techniques. In this final part of our Power BI blog series, we cover how AI plays a role in the Power BI service and the future of AI in Power BI products.
What is the Power BI service?
Power BI is a collection of software services, apps, and connectors that work together to turn your unrelated sources of data into coherent, visually immersive and interactive insights. The Power BI service is one of the apps that makes up the entire Power BI platform. The Power BI service is an online Software-as-a-Service (SaaS) which means that the infrastructure, hosting and upgrades are all taken care of by the provider – Microsoft. The Power BI service is the location where non-report authors will interact with reports, dashboards and analysis. Within the Power BI service, AI is used to make navigation clearer, help users gain better insights from their data and keep their deployment as engaging and tidy as possible.
Power BI service – Quick Insights
Quick Insights is a function with the Power BI service that utilises AI to interrogate your dataset and present visualisations that will provide greater insight. Quick Insights automatically searches datasets uploaded to Power BI for hidden insights. The types of insights derived can be broken down into the below types:
Major factor(s)
Finds cases where a majority of a total value can be attributed to a single factor when broken down by another dimension.
Category outliers (top/bottom)
Highlights cases where, for a measure in the model, one or two members of a dimension have much larger values than other members of the dimension.
Time series outliers
For data across a time series, detects when there are specific dates or times with values significantly different than the other date/time values.
Overall trends in time series
Detects upward or downward trends in time-series data.
Seasonality in time series
Finds periodic patterns in time-series data, such as weekly, monthly, or yearly seasonality.
Steady Share
Highlights cases where there is a parent-child correlation between the share of a child value in relation to the overall value of the parent across a continuous variable.
Correlation
Detects cases where multiple measures show a correlation between each other when plotted against a dimension in the dataset.
View Related
Within the Power BI service, AI is also working to improve productivity and usability in the background. Within each workspace in the service, users can find related reports to the one they are currently looking at. This functionality interrogates the dataset and compares it to the datasets elsewhere in the service. Where there is cross-over, the service will suggest the user look at the connected reports. This only suggests content that the user is authorised to see, and this will still uphold security in place throughout the service, workspace and individual reports.
One issue that occurs over time, in any self-serve reporting environment, is duplication. Many users, each capable and able to create reports, will no doubt create similar insights in their departments or workspaces. However, with the ‘view related’ function, Power BI users can mitigate duplication, prevent unnecessary bloating on the service and most importantly avoid duplication of time by finding the report they need without having to build it again.
This feature can also have a significant impact on housekeeping. The related content source may be a dashboard – perhaps you want to understand all the reports feeding a specific dashboard? ‘View related’ makes this exercise simple.
Power BI Roadmap
The upcoming Power BI roadmap is packed with AI-driven functionality. One of the four tracks of the roadmap is ‘Pervasive Artificial Intelligence for Power BI’. This commitment from Microsoft shows the value in AI-driven analysis for Power BI users.
If you would like to find out more about Power BI visit our Power BI page. For more information on our AI and data science, visit our Data Science page.