BI and Analytics with SQL Server and R - Tomaz Kastrun
This one-day pre-con is for Data Analyst and business people who would like to learn more on SQL Server and R integration, what R has to offer and what are practices for creating the environment to start working on statistics and predictive analytics.
It will be demo based and users will be able to play along. It is encouraging to bring your own laptop and code along. You will need a laptop running SQL Server 2016 (Enterprise or Developer edition) and Visual studio 2015 Community edition (with RTVS installed). Script will be available upon session; datasets will be distributed on session (Adventureworks/WideWorldsImporters/Sales or CRM data).
Participants will walk away with understanding how R in SQL Server works, basic knowledge on R Language, manipulating data with R packages, methods and algorithms for statistical analysis, predictive analytics available in RevoscaleR package and visualizing data.
Materials will be handed as R and T-SQL Code and PPT Slides.
Breaks and lunch will be defined by organizers and later agreed with group of participants. Maximum number of participants is decided by organizer.
Language Session: English
Level: 200-300
Module 1: 9.00 - 10.00 Introduction to powerful duo R and SQL Server – Introduction to R, Integration to R and SQL Server, architecture and how to efficiently use both. (Module: 70% presentation; 30% Demo; 0% hands-on)(10 min break)Module 2: 10:10 – 11.20 Knowing R – we will cover some basics on R language, using data frames, lists, importing and exporting data and covering different libraries. (Module: 30% presentation; 70% Demo/hands-on)(10 min break)Module 3: 11.30 – 12.30 Knowing statistics - from general statistics to multivariate statistics for data analysing and complexity reduction. Algorithms for prediction analytics (Decision trees, Naive Bayes, Regression) and text mining. (Module: 60% presentation; 40% Demo/hands-on)(30 min Lunch break)Module 4: 13.00 – 14.20 Using RevoScaleR Package – We will cover what RevoscaleR Package is offering for parallel and distributed workload between R Engine and SQL Server. We will perform different algorithms for statistics and for predictive analytics and compare them to previous approaches. (Module: 50% presentation; 50% Demo/hands-on).(10 min break)Module 5: 14.30 – 15.15 Visualizing Data - Data preparation, data visualization. Using SSRS, Power BI, R libraries and third party tools for data presentation. Covered by this module will also be vertical presentation of data flow in a company and how (Module: 50% presentation; 50% Demo/hands-on)(10 min break)Module 6: 15.25 – 16.00 Story completing – cases from the field. Using CRM Data and predictions within SQL Server, Creating and using Basket analysis, Classification of customers to monitoring and performance analysis for DBA and system administrators. (Module: 100% Demo)
About the Speaker : Tomaž Kaštrun
Tomaž Kaštrun (t:@tomaz_tsql) is BI / DEV developer and data analyst. With more than 15 years of experiences in the field of databases, business warehouses and development, with focus on T-SQL programming and query optimization. Besides BI development, his focus is also also data mining, statistics and research.
He has been working with Microsoft SQL Server since version 2000. He is also Microsoft Certified Professional, Microsoft trainer and Microsoft Data Platform MVP.
Tomaž is a blogger, author of many articles, speaker at various community and Microsoft events and an avid coffee drinker.
Currently working at SPAR Business Services GmbH .
