<img height="1" width="1" src="https://www.facebook.com/tr?id=152309752858854&amp;ev=PageView &amp;noscript=1">

Power BI: Top 5 no-code AI & machine learning features

With 850,000 active users and more than 30 million dashboards and reports hosted, Power BI has fast become one of the most versatile self-service business intelligence tools for no-code data analysis.

Microsoft has consistently enhanced Power BI’s user-friendly capabilities in 2019, and one huge area of improvement that enterprises need to start taking advantage of is the rapid introduction of artificial intelligence (AI) capabilities in Power BI's dashboards and reports, which are key to unlocking deeper, previously unseen insights from our data.

Since the introduction of image and text analytics in March and AI services integration in June, Power BI users have been able to leverage the latest low-code conversational AI, natural language processing and voice recognition features for their own business data and extract greater actionable insights - all without having to be experts in the data science behind it all. Combined with scalable compute and more sophisticated data modelling, AI is no longer an inaccessible tool for regular analysts - it's already here,

We have rounded-up the top 5 best AI capabilities in Power BI that you need to start leveraging in your data analysis. Keep this page bookmarked as we update the list with Power BI's latest capabilities.

#1 - Power BI Q&A: Natural Language Processing

Image source: Microsoft

Data is more interesting to explore - and fun -  when it’s interactive and responsive.

By leveraging natural language processing (NLP), Power BI’s Q&A feature offers new ways to explore data and find hidden insights - and has received tons of fresh updates.

Power BI Q&A lets regular business users explore their datasets by asking specific questions, and responding with the best and most relevant answers, fully visualised. It also displays suggestions to help you form your question, and each visualisation changes dynamically as you modify the question. You can even set how Q&A displays answers; for example, asking to ‘show monthly performance as a graph’; this lets your users drill down into the results their way.

Power BI’s NLP capabilities have continued to evolve since its release. It recently enabled the in-built Q&A feature to be trained like a machine learning model, adapting to and understanding company-specific language and phrases to provide users with relevant results more intuitively. Report authors can also now view every single natural-language query asked by users so they can fine-tune how Power BI responds next time, improving its accuracy and responsiveness.

Perhaps in response to Tableau’s popular Ask Data feature set, Power BI’s natural language query and existing visual and text analysis components will be newly bolstered with updates that introduce better text and handwriting recognition, and entity detection. With these improvements, it’s easily one of the most polished NLP business intelligence services on the market across all of its different versions: Power BI Desktop, Power BI Pro and Power BI Premium.

It's important to note that Power BI Q&A only provides results about the data in Power BI, so it's only as extensive in capability as the data that is provided. Currently, only natural language queries asked in English are supported, but there are other languages such as Spanish currently in preview.

Version Availability: Power BI Q&A is available with a Power BI Pro or Premium licenses only.


#2 - Siri voice integration with Power BI Mobile: Conversational AI


Power BI offers on-the-go versions of its analytics reporting via a set of mobile applications for Android, iOS and Windows 10 smartphones and tablets - even Apple Watch.

On mobile, users can connect to and consume predefined dashboards and reports created in Power BI Desktop and on-premises reports on Power BI Report Server easily.

However, Power BI iOS mobile users are extra lucky and at the moment have exclusive access to two nifty conversational AI capabilities built into the app.

For starters, you can now add Siri shortcuts to frequently viewed dashboards and reports, enabling faster, seamless access to important items. You can set and use a specific voice command of your choice to immediately view and consume the data you want to explore, without sifting through the lists.

Power BI iOS users can also ask specific questions about their data and gain new insights and suggestions using the Q&A virtual analyst. The unique chat feature is in-built and tailor-made for mobile. The natural language querying is accessed via the action menu in each dashboard, where you can either type your question in the chat, or use iOS native speech recognition (voice) to ask questions.

For example:

  • While viewing a retail sales analysis report, you can ask Power BI Q&A to specifically filter last year’s sales against current performance and visualise it in a bar chart optimised for the smaller screens of your smartphone or tablet.
  • You can also use your voice via your mobile device’s microphone to ask Power BI to visualise the number of sales made in a given month in a column chart for a more detailed exploration - and for a more immediate answer.

The Power BI mobile Q&A virtual analyst is still evolving and the Siri integration is fresh out of this year’s Microsoft Business Application Summit, but both are great features and steps towards integrated AI capabilities that definitely should be leveraged by all Power BI users.

Version Availability: Siri voice integration and Q&A virtual analyst are only available on Power BI Mobile iOS app for iPad, iPhone and iPod Touch with iOS 10 or later.

#3 - Azure Cognitive Services integration: Machine Learning (ML)



Azure Cognitive Services (ACS) is now integrated with Power BI Premium, opening up the door for users to leverage pre-trained machine learning models for greater insights. 

With ACS (no subscription required), business-level Power BI users can apply different ML algorithms to enrich their data during the self-service data preparation for data flows, which is the process of ingesting, cleansing, transforming, integrating and curating data from numerous sources in the business

Power BI now stores these ACS and ML-enhanced data flows in Azure Data Lake Storage, providing broader access to data engineers and data scientists throughout your organisation to leverage more advanced tools like Azure Databricks, Azure Machine Learning and Azure SQL Data Warehouse and apply additional advanced analytics and AI-driven capabilities to datasets.

Currently, there are four intelligent pre-trained models we can apply in Power BI to our text-based and image-based datasets:

  • Language detection
  • Key phrase extraction
  • Image tagging
  • Sentiment scoring.

Users can now use these machine learning models within Power BI  (and the AI Insights Browser in Power Query Online) to extract insights, for example from images by detecting relevant objects, or analyse text fields like customer feedback to pinpoint important phrases and positive comments for future analysis, or compare sentiment in different languages. All of these AI-powered benefits come back to our end-users who consume Power BI reports the most.

Version Availability: Azure Cognitive Services integration is available with a Power BI Premium licenses only.



# 4 - Power BI Key Influencers Visual: AI-powered visualisation

Key Influencers Visual - Top 5 AI features in Power BI

Image source: Microsoft

The Key Influencers Visual is a handy visualisation option in Power BI that helps users understand the factors that drive metrics (categorical and numerical) they’re most interested in, and how groups of key influential factors affect the selected condition. It’s fully interactive, and you can explore data using slicers, filters and other visualisations to represent the results.

Using a Key Influencers Visual to analyse your data sets means you can identify and contrast the relative importance of influential factors, and which have the most relevance. It’s entirely AI-powered, with intelligence built-in that runs behind the scenes to help users find new insights.

Power BI’s native integration with Azure Data Services means users can actively leverage substantial AI capabilities like machine learning to run regression analysis to model the influencer data and segmentation of the data overall - just by using the Key Influencers Visual.

While the public preview feature is still evolving and limited by a lack of consumption support for Power BI Embedded or Power BI Mobile and no support for metrics that are aggregate and measures, Microsoft have recently announced they are releasing two new companion AI visuals, Distribution Changes and Decomposition Tree, that further bolster Power BI.

  • Distribution Change: Will analyse what makes a distribution look different.
  • Decomposition Tree: Will enable users to drill into any dimension to further understand what influencers drive a key metric in question.

Ultimately, Power BI receiving more AI-enhanced visuals is only a great thing for deeper exploration and delivering previously unseen insights for both data analysts and business users.

Version Availability: Key Influencers Visual is available with a Power BI Premium or Pro licenses only.


# 5 - Advanced AI modelling with Azure Machine Learning


Azure Machine Learning (AML) is a low-code toolset that enables users to build entire data models, machine learning algorithms, pre-processing modules and more components through drag and drop gestures on an interactive design surface. You can run training experiments, examine results, link them together graphically and then deploy them to services like Power BI.

The latest Azure ML and Power BI integration empowers data scientists to more easily export their customised ML training models directly to Power BI (Premium only), helping business users take better advantage of the automated machine learning (AutoML) feature, which:

  • Enables business analysts using Power BI to use their dataflows and choose the best model to drive desired outcomes
  • Train their model on their data within Power BI and without needing to code anything
  • Automated reporting on its performance using Power BI’s dashboard and visualisations.

AutoML then allows users to evaluate and customise their best ML models until they are optimised, and then apply them to future datasets for predictive insights.

The best part? Most of the advanced data science aspects of ML modelling is automated and managed by Power BI, meaning both advanced data scientists and non-technical users can explore and benefit from AutoML’s capabilities - which continue to improve with regular updates.

As of June 2019, models created in Power BI can also be exported to Azure ML.

Version Availability: Azure Machine Learning integration is available with a Power BI Premium or Pro licenses only.



The key to taking advantage of Power BI's expansive and evolving AI and machine learning capabilities is a strong foundational business intelligence strategy. Before you begin exploring Power BI's many features, read our free-to-download BI whitepaper for a step-by-step guide on how to deliver better insights from your analytics.

4 Best Practices for Successful Business Intelligence


Tags: Data Analytics, Power BI, Artificial Intelligence