Automated Machine Learning (AutoML) in Power BI

There are a lot of features in Power BI that allow the business analyst to enrich their solutions with Artificial Intelligence. In this session you will see how you can create a Machine Learning model directly in Power BI using the AutoML capability.

Transcript

1. #GlobalAIBootcamp Automated Machine Learning (AutoML) in Power BI Dave Ruijter Solution Architect Macaw Netherlands inbox@daveruijter.nl @DaveRuijter linkedin.com/in/DaveRuijter ModernData.ai
2. #GlobalAIBootcamp
3. #GlobalAIBootcamp What will we cover AI in Power BI Analyst toolkit Automated ML Cognitive services How to start End User Insights AI visualizations Natural language capabilities
4. #GlobalAIBootcamp AI in Power BI
5. #GlobalAIBootcamp Power BI advanced capabilities 2015 2019 2016 2018 Desktop Run R in Power BI visuals Desktop R in Power Query Desktop Phyton in Visuals + Power Query AI/ML features - AutoML in dataflows - Azure ML integration - Cognitive Services integration
6. #GlobalAIBootcamp AI in Power BI Sentiment Analysis Key Phrase Extraction Create ML models Explore predictions Python Integration R Integration Extend with Azure ML Integrate into reports AI Visualizations Natural Language
7. #GlobalAIBootcamp AI in Power BI Sentiment Analysis Key Phrase Extraction Create ML models Explore predictions Python Integration R Integration Extend with Azure ML Integrate into reports AI Visualizations Natural Language
8. #GlobalAIBootcamp Azure Machine Learning
9. #GlobalAIBootcamp demo time! because life is boring without risks
10. #GlobalAIBootcamp Key Takeaways – AutoML Power BI + Automated ML enables you to:  Automate most of the standard machine learning practices in a non- intrusive yet transparent way  AI in Power BI is not only about achieving great accuracy but about easy translating this accuracy to business impact
11. #GlobalAIBootcamp Key Takeaways – AutoML Automated ML in Power BI Automated ML in Azure Designed for the business analyst/BI professional Designed for for the data scientist/data engineer Skillset required is Power Query and Power BI dataflows (OOB point and click functions, M code) Skillset required is Azure, Python and Jupyter notebooks ML model lives completely inside Power BI and can only be used by Power BI ML model lives in Azure and can be consumed as a webservice by lots of different applications (including Power BI) Requires Power BI Premium and utilizes capacity Requires an Azure subscription and utilizes Azure consumption
12. #GlobalAIBootcamp AI in Power BI Sentiment Analysis Key Phrase Extraction Create ML models Explore predictions Python Integration R Integration Extend with Azure ML Integrate into reports AI Visualizations Natural Language
13. Cognitive services
14. #GlobalAIBootcamp demo time! because life is boring without risks
15. #GlobalAIBootcamp Key Takeaways – Cognitive Services Cognitive services in Power BI Cognitive services in Azure Designed for the business analyst/BI professional Designed for for the data scientist/data engineer Skillset required is Power Query and Power BI dataflows (OOB point and click functions, M code) Skillset required is Azure, coding language (C#, Java, Node.js, Python) Lives completely inside Power BI and can only be used by Power BI Live in Azure and can be consumed as a webservice by lots of different applications Require Power BI Premium and utilizes capacity Require an Azure subscription and utilizes Azure consumption
16. #GlobalAIBootcamp AutoML / Cognitive Services in Power BI: how to start? 1. Power BI environment 2. Pro Licens (trial) 3. Power BI Premium (Azure version for dev/test) 4. Get your data in a dataflow -> create a model 5. Get your data in desktop -> invoke Cognitive Services • Download the latest Power BI Desktop to try AI Visuals https://powerbi.microsoft.com/en-us/desktop/ • Quick start guides to get started with Automated ML, Cognitive Services and Azure ML: https://docs.microsoft.com/en-us/power-bi/service-machine-learning-automated
17. #GlobalAIBootcamp Session takeaways Power BI + AI:  Gives analysts the opportunity to enrich their data using AI and build their own machine learning model without writing any code  Extension points for data scientists to plug in their Python/R scripts or to bring in machine learning models they created elsewhere  Findings insights seamlessly and interacting with data in a natural way
18. #GlobalAIBootcamp Download for free! Https://bit.ly/cheatsheetpbi Dave Ruijter Solution Architect Macaw Netherlands inbox@daveruijter.nl @DaveRuijter linkedin.com/in/DaveRuijter ModernData.ai