Machine Learning in R—–NYC class offering

SupStat is offering a 5-day intensive course in machine learning techniques in R starting this Sunday, October 5th at its Times Square office. Classes are held on October 5th, 12th, and 18th, and November 2nd and 9th, from 10 am to 5 pm for total 35 hours of classroom time and 3 weeks of projects . If you’re new to machine learning, here’s what it’s all about:

You can view the

Syllabus at r-programming-intensive-intermediate

Testimonial at: testimonials

Students’ work at students-work

Students’ projects with video recording at demo day

Machine Learning (alternatively referred to as statistical learning, pattern classification, etc.) is a broad field which seeks to simulate real-world processes by way of computational modeling. In some cases we start with a real-world outcome (stock price data, for example) and a set of factors potentially involved in producing that outcome, and explore models which will find meaningful relationships between the two. In essence we want the machine to learn how the underlying process works so that we can then use it to predict future outcomes. Other times we have no knowledge of cause or effect but simply have a wide array of data we’re looking for patterns in (social media behavior, for example).

Machine Learning is everywhere! It’s behind the sidebar generating targeted ads on your Facebook page, it’s in your smart phone app that acts on your voice commands, and it even works behind the scenes to elect candidates in some of our highest public offices. And yet despite this complex landscape of problems machine learning is trying to solve, there is one open-source tool, R, that can meet nearly all of them. After our course you’ll walk away with not only the knowledge of statistical methods for addressing these challenges, but you’ll have the computational skills to implement them!

Our course will cover the following major topics:

• Foundations of Data Mining
• Performance Measures and Dimension Reduction
• KNN and Naïve Bayes Models
• Tree Models and Support Vector Machines
• The Association Rule and Domain-specific Models

We have had people who have no programming experience and took last math class in high school and turned out to be top students in our class! We also have had CMU alumni of computer science, engineer of Google, managers of analytic teams, experienced software engineers to take the class with us in the past. The key to do well in class and after class is to try hard, really really hard to learn the skills and apply them into real world cases.

Interested students should be comfortable with the fundamental structures and programming elements in R. Please visit our course page to sign up, and email contact@supstat.com if you have any questions about the course or would like to find out more about our other offerings.