Spsnyc 2016 machine learning
-
Upload
fernando-leitzelar-mba-pmp -
Category
Software
-
view
94 -
download
0
Transcript of Spsnyc 2016 machine learning
![Page 1: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/1.jpg)
Hottest Buzz Out There: Integrating Predictive Analytics, SharePoint and Azure Machine Learning
Fernando Leitzelar, PMP Vice President ITSM
![Page 2: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/2.jpg)
THANK YOUEVENT SPONSORS
We appreciated you supporting the
New York SharePoint Community!
• Diamond, Platinum, Gold, & Silver have tables scattered throughout
• Please visit them and inquire about their products & services
• To be eligible for prizes make sure to get your bingo card stamped by ALL sponsors
• Raffle at the end of the day and you must be present to win!
![Page 3: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/3.jpg)
CONFERENCE MATERIALS
• Slides / Demo will be posted on Lanyrd.com• http://lanyrd.com/2016/spsnyc
• Photos posted to our Facebook page
• https://www.facebook.com/sharepointsaturdaynyc
• Tweet Us - @SPSNYC or #SPSNYC
• Sign Up for our NO SPAM mailing list for all conference news & announcements
• http://goo.gl/7WzmPW
• Problems / Questions / Complaints / Suggestions• [email protected]
![Page 4: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/4.jpg)
• Visit ExtaCloud’s booth for wrist bands!
Scallywag's Irish Pub
508 9th Ave, between 38th & 39th. [6 minutes walk]
Scallywags also serves food.http://www.scallywagsnyc.com/
![Page 5: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/5.jpg)
SpeakerFernando Leitzelar, PMP
Vice President ITSM
Fernando Leitzelar is a senior SharePoint Evangelist and Vice-president with a Large Bank As a consultant he regularly interfaced with clients and development teams to design SharePoint-based solutions. Fernando has progressively held SharePoint positions ranging from developer and administrator to Architect and Manager. He has been a SharePoint Saturday Speaker since 2010, having worked extensively on designing and architecting sophisticated SharePoint based applications. He maintains expertise in Office 365, Azure, SharePoint 2016/2013/2010/2007/2003, BI and Machine Learning Solutions.
Twitter: @fleitzelar
Blog: http://sharepointusa.wordpress.com
![Page 6: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/6.jpg)
• What and How ?Introduction to ML and
Predictive Analytics
• Predictive Analytics using Machine LearningPredictive Analytics
• Machine Learning Studio• Building ML models• Create a Web Service
Azure Machine Learning
• Consume Machine Learning ModelSharePoint Online
Agenda
![Page 7: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/7.jpg)
Predictive AnalyticsPredicting future performance based on historical data
![Page 8: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/8.jpg)
Predictive analytics encompasses a variety of statistical techniques from predictive modeling, machine learning, and data mining that analyze current and historical facts to make predictions about future or otherwise unknown events.
![Page 9: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/9.jpg)
ADVANCED ANALYTICSBEYOND BUSINESS INTELLIGENCE
![Page 10: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/10.jpg)
From Descriptive to Prescriptive
Analytics Maturity Level
![Page 11: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/11.jpg)
What happened ?• Reporting:
Statistics
Why did it happen ?• Analysis: Excel, OLAP
What is happening ?• Monitoring: Dashboards,
Scorecards
What will happen ?• Prediction: Data Mining,
Machine Learning
Evolution of Predictive Analytics
2000s
1990s
1980s
2010s
![Page 12: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/12.jpg)
Machine LearningComputer Systems that improve with experience
![Page 13: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/13.jpg)
CLASSES OF LEARNING PROBLEMS• Classification: Assign a category to each item (Chinese | French | Indian | Italian |
Japanese restaurant).
• Regression: Predict a real value for each item (stock/currency value, temperature).
• Ranking: Order items according to some criterion (web search results relevant to a user query).
• Clustering: Partition items into homogeneous groups (clustering twitter posts by topic).
• Dimensionality reduction: Transform an initial representation of items into a lower-dimensional representation while preserving some properties (preprocessing of digital images).
![Page 14: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/14.jpg)
WHAT IS MACHINE LEARNING?
Methods and Systems that …
Adapt based on recorded
data
Predict new data based on recorded
data
Optimize an action given
a utility function
Extracthidden
structure from the
data
Summarizedata into concise
descriptions
![Page 15: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/15.jpg)
MACHINE LEARNING IS NOT
Methods and Systems that …
can yield Garbage-In-Knowledge-
Out
perform good predictions without data modeling &
feature engineering
Silver-bullet for all data-
driven tasks –it’s a powerful
data tool!
are a replacement for business rules – they
augment them!
![Page 16: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/16.jpg)
TRANSFORMATIONAL
![Page 17: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/17.jpg)
A Good Machine Learning Tool would allows us to
solve extremely hard problems betterextract more value from Big Data
approach human intelligence
drive a shift in business analytics
![Page 18: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/18.jpg)
Data Science is far too complex today• Access to quality ML algorithms, cost is high.• Must learn multiple tools to go end2end,
from data acquisition, cleaning and prep,machine learning, and experimentation.
• Ability to put a model into production.
This must get simpler, it simply won’t scale!
PROBLEMS ML NEEDS TO ADDRESS …
![Page 19: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/19.jpg)
![Page 20: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/20.jpg)
PREDICTIVE ANALYTICS AND ML SCENARIOS
Predictive maintenance
![Page 21: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/21.jpg)
Classification
Regression
Clustering
Anomaly Detection
AZURE ML ALGORITHMS
![Page 22: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/22.jpg)
ADVANCED ANALYTICS TODAYHARD-TO-REACH SOLUTIONS
Break away from industry limitations
![Page 23: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/23.jpg)
WHAT HAS IT GOT TO DO WITH SHAREPOINT ?
MLStudio
API
M
![Page 24: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/24.jpg)
AZURE MACHINE LEARNING
![Page 25: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/25.jpg)
AZURE MACHINE LEARNING
Azure Portal
ML Studio
ML API Service
Operational Team
Data Scientists and Data Professionals
Software Developers
![Page 26: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/26.jpg)
AZURE ML STUDIO
![Page 27: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/27.jpg)
MICROSOFT AZURE MACHINE LEARNINGBuilt for a cloud-first, mobile-first world
![Page 28: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/28.jpg)
Step 1
• Data Preparation and Feature Engineering
Step 2
• Train and Evaluate Model
Step 3
• Deploy Web Service
BUILDING ML MODEL
![Page 29: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/29.jpg)
![Page 30: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/30.jpg)
RECEIVER OPERATING CHARACTERISTIC CURVE
![Page 31: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/31.jpg)
ROC CURVE
![Page 32: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/32.jpg)
ROC CURVE
![Page 33: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/33.jpg)
Reduce complexity to broaden participationMICROSOFT AZURE MACHINE LEARNING
FEATURES AND BENEFITS
• Accessible through a web browser, no software to install;
• Collaborative work with anyone, anywhere via Azure workspace
• Visual composition with end2end support for data science workflow;
• Best in class ML algorithms;• Extensible, support for R.
![Page 34: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/34.jpg)
MICROSOFT AZURE MACHINE LEARNING
FEATURES AND BENEFITSRapid experimentation to create a better modelImmutable library of models, search discover and reuse;Rapidly try a range of features, ML algorithms and modeling strategies;Quickly deploy model as Azure web service to our ML API service.
![Page 35: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/35.jpg)
• https://azure.microsoft.com/en-us/documentation/articles/machine-learning-algorithm-choice/
• https://azure.microsoft.com/en-us/documentation/services/machine-learning/
• Azure Machine Learning Essentials Book
• https://channel9.msdn.com/blogs/Cloud-and-Enterprise-Premium/Building-Predictive-Maintenance-Solutions-with-Azure-Machine-Learning
• Channel 9
AZURE MACHINE LEARNING RESOURCES
![Page 36: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/36.jpg)
KEY CONCEPTSData
Model
Parameters
Learning Prediction
Decision Making
Utility Function
![Page 37: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/37.jpg)
STEPS TO BUILD A MACHINE LEARNING SOLUTION
![Page 38: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/38.jpg)
AZURE DATA MARKET ML APPLICATIONS
• http://text-analytics-demo.azurewebsites.net/
• https://churn.cloudapp.net
• http://how-old.net/#
![Page 39: Spsnyc 2016 machine learning](https://reader031.fdocuments.net/reader031/viewer/2022030305/5872eb851a28abfa548b71c9/html5/thumbnails/39.jpg)