Time: September 12, 2014 from 9am to 11:45am
Location: SAS Netherlands
Street: Flevolaan 69
Website or Map: http://www.sas.com/apps/wtrai…
Event Type: master, class, seminar
Organized By: SAS Opleidingen
Latest Activity: Jul 16, 2014
Data Scientist was called ‘The Most Sexy Job of the 21st Century’ in a 2012 Harvard Business Review article. The first questions that this session answers are: what is a data scientist, and what skills and competencies should a data scientist have? Which areas of expertise are relevant if you want to become a data scientist? This one day overview serves as a kickstarter for the aspiring analyst who wants to become a Data Scientist and teaches you which combination of social, analytical and data skills is essential to be successful in a data science role.
Two key elements stand out in terms of skills: statistics and data ‘munging’. The latter is a combination of techniques for preparing your data set for analysis; the former is needed for various purposes. Data is the foundation for any analysis and can be sourced from a wide variety of platforms and applications, each in its own form and format. We’ll quickly cover the most common scenarios for obtaining data, and explain which types of (big) data platforms can be used to run your analysis on. Just getting data is not enough though; understanding it is far more important. Which type of variables to use, how to select the right variables and which role they play in an analysis.
A basic understanding of statistics is essential in any data science project, so we’ll help you get started by building the basic statistical knowledge needed for your projects. The simplest form of statistics is Descriptive Statistics which helps you in understanding your data and assessing the quality of the data set(s) you’re going to work with. Next, there’s exploratory analysis to find known and previously unknown relationships and patterns in your data. We will also explain the other types of statistical analysis, like inferential, predictive, and causal analysis.
We continue our tour with an overview of data mining techniques who are the real work horses of the data science world. You’ll learn which techniques are available, what their purpose is and which algorithm to use for which data science challenge. What you do with these algorithms is building a model; some models are better suited for a particular task or data set than others, and you’ll learn how to evaluate and compare them.
Finally you’re going to discover how to visualize and present your findings by using various visualization techniques. Correctly visualizing your data is the icing on the cake and helps you to convey your message to a non-technical audience.