Power BI uses Natural Language to move from Data to Answers

Power BI uses Natural Language to move from Data to Answers

This article was originally published on LinkedIn Pulse by our Data and AI Practice Lead, Tarun Agarwal.

Microsoft recently revamped Q&A functionality in Power BI. The newly released provides an intuitive and interactive way to consume business intelligence. Users can naturally ask questions like Google search and get answers in BI report visuals.

Usually, when we think about machine learning and artificial intelligence, we think about prediction, forecast, and actions. However, AI features can also be used in historical reporting. Users can get answers about “what happened” or “why it happened” using Power BI features. Microsoft is infusing AI in every step of BI. We do not need Data Scientists to leverage AI features. Business analysts and end-users can also use AI features in BI reports.

Users cannot be productive by going through multiple pages of reports to answer business questions. Analysts cannot also anticipate all possible questions a user might have. Therefore, even while they have BI reporting, users still struggle to find answers to questions that come up. End users will be so much more empowered if they can ask business questions naturally as they do on search engines like Google and Bing. 

Power BI Q&A visual allows users to ask the question in natural language and get answers in report visuals. Q&A visual is also integrated with other report visuals and work seamlessly. The Q&A visual will interact with other report visuals, and slicers the way Power BI works.

Power BI is using natural language processing techniques behind the scenes to link questions with data, understand the key phrases, intent, KPI. Q&A visual uses synonyms and annotations to answer different types of questions a user can ask.

Q&A visual also provide “teach Q&A” functionality where it uses Q&A to teach Q&A visual. Microsoft has also implemented a comprehensive feedback loop to improve answering different types of questions from different users. Report authors can view all questions users asked and identify the questions Q&A visual could not understand. The questions can be fixed by teaching Q&A using synonyms.

Microsoft has been continuously working to make its products intelligent and provide tangible business value to the customers. Power BI is infusing various AI features that are pervasive and intuitive. Power BI is democratizing machine learning and artificial intelligence to putting in hands-on end users to unlock powerful business insights.