Ai in power platform
Download as PPTX, PDF1 like527 views
The AI in PowerPlatform presentation September Meetup Vancouver PowerBI and Excel User Group. Talk by Joseph and myself
Ad
More Related Content
What's hot (20)
Similar to Ai in power platform (20)
Ad
More from Berkovich Consulting (20)
Ad
Recently uploaded (20)
Ai in power platform
- 1. Yana Berkovich Data Platform MVP @yana_Berkovich YanaBerkovich.com AI in PowerBI Vancouver PBI User Group Joseph Yeates, @FthrsAnalytics https://feathersanalytics.com/
- 2. AI in Power BI & Power Platform Joseph Yeates, aCAP www.feathersanalytics.com @FthrsAnalytics in/josephyeates237
- 3. Yana Berkovich BI Analyst& DEV, Microsoft Data Platform MVP Consultant, Product Manager BusinessIntelligenceand Data Visualization Consultant https://www.linkedin.com/in/yanaberkovich http://yanaberkovich.com @Yana_Berkovich
- 4. AGENDA • What is AI and analytics • Natural language Query • Simple prediction algorithm in existing visuals • The ML model • AI builder and image recognition • Summary
- 6. Algorithms Artificial Intelligence Machine Learning Human Input & Interpretation is Always Required What is Predictive Analytics?
- 7. WHAT IS AN ALGORITHM? An algorithm is a set of rules to be followed in problem solving operations Initial Input Computation Instructions Produce Output
- 8. SORTING ALGORITHM EXAMPLE Sorting algorithm takes an unsorted list, applies pre- determined instructions and produces a sorted list Unsorted List Sorting Instructions Sorted List
- 9. WHAT IS ARTIFICIAL INTELLIGENCE? Artificial Intelligence is “intelligence” demonstrated by machines and is often associated with cognitive function
- 10. WHAT IS ARTIFICIAL INTELLIGENCE? Artificial Intelligence is “intelligence” demonstrated by machines
- 11. TYPES OF ARTIFICIAL INTELLIGENCE Artificial intelligence applied to a specific task Applied AI General AI Ability to accomplish a variety of cognitive tasks
- 12. DIFFERENT TYPES OF ANALYTICS • AI is in the Predictive analytics space, we are predicting based on the past the future observations
- 13. WHAT IS MACHINE LEARNING? Machine Learning is an application of Artificial Intelligence that learns to construct Algorithms from experience
- 15. Lifeexp by year line by country Pop by continent column sort by pop Average gdpPercap by average lifeExp scatter by country legend continent Gdppercap by top 10 country in year 2007 ASK A QUESTION
- 16. LET’S GO TO POWERBI • Quick Insights
- 17. THE FIRST STEP – EXPLORING AND VISUALIZING YOUR DATA – QUICK INSIGHTS Quick insights mode Quick insights mode
- 18. ADDING INFORMATION TO OUR DATA VISUALIZATION PowerBI Min value line Max value line Trend line – we can see that the AVG delay time increases? Expediential Smooth Seasonality – 7 points (week in a month) Ignore last 10 points – to check our prediction Forecast length- to see what the other 7 days will look like
- 19. STATISTICS…(YES WE KINDA HAVE TO) WHICH PREDICTION MODEL ARE WE USING? Average Single Exponential Smoothing Exponential smoothing is a rule of thumb technique for smoothing time series data using the exponentialwindow function. Whereas in the simple moving average the past observations weighted equally,exponential functions are used to assignexponentially decreasing weights over time. ( Wikipedia to the rescue… ) Moving Average The last month might be a better prediction for flights than the last 20 months Weighted Moving Average Some observations are more significant than others, flights of a domastic flight company have different performance and cannot be compared to others or big vs small planes Can be chosen, for the single smoothing, between 0.1 and 0.9, is chosen through a local optimal minimum value We choose the best value for α so the value which results in the smallest MSE. (Mean of Square Errors)
- 20. ADDING INFORMATION TO OUR DATA VISUALIZATION PowerBI – How can we explain the predicted results? Trend line – we can see that the AVG delay time increases? Possible explanations for this prediction and what can we do about it – Prescriptive Analytics • End of October - Thanksgiving? • Weather changes at the airports for the worse • The trend line doesn’t continue for the predicted data • How can we control the Alpha? Well in Power View for O365, not in PowerBI yet
- 21. POWERBI AI
- 22. IN AZURE ML
- 23. POWER PLATFORM
- 24. THE AI BUILDER - POWERAPPS PowerBI is no longer a “single in the mingle” it has a family – the power platform So How do we do AI with the PowerPlatform? Meet the AI builder
- 25. THE AI BUILDER HOW TO BUILD YOUR ON AI? Create the terms in CDM Create new model Select type - image Add picturesAdd TagsTrain Save Test Add to an application
- 26. CDM COMMON DATA MODEL – ADDING ENTITIES TO TAG
- 27. WHY DOES AN AI MODEL NEEDS TO TRAIN? Search results of my pictures in google for the word: “BIRD”
- 29. TEST - WHAT IS THE PREDICTION ACCURACY?
- 30. SHARING YOUR APPLICATION – SHARE THE BIRD
- 31. AND THE RESULT? With Ken’s help, 1 Goose was successfully detected in the picture!
- 32. BUT WAIT…GOOGLE CAN DETECT MY MOM IN EVERY PICTURE SO WHAT’S SO EXCITING ABOUT THIS GOOSE CHASE? • You can add the AI feature right into your own application • Building inspection object detection in the application • Business Data and Business Pictures • One Drive for Business or Azure repository and files management • CDM – the tagging is an object in your Common Data Model not stand alone object
Editor's Notes
- #3: I am a Data Analyst at BlueShore Financial in North Vancouver, British Columbia and hold my Certified Analytics Professional Designation. I use SQL Server, R Studio, and Power Query in Excel & Power BI to model, manipulate, analyze, and visualize data. I also run internal information sessions on predictive analytics and training sessions on Power Query and Power BI. I’m also a member and regularly speak at the Vancouver Power BI User Group. My website is Feathers Analytics .com, where I blog about all things Power BI and R but most importantly – how to bring the different strengths of these tools together to build powerful solutions that provide actionable insights.
- #5: – Joseph adds his slides – Joseph makes introduction and adds the slides Yana to add the data flow and prediction model for aiplanes and not bike buyers What is a ML model – Yana to add a slide Slide about premium Joseph to add all the AI