Advanced masterclass 2: Random forests

Advanced masterclass 2: Random forests

2,540.00

MELBOURNE: 29-30 April

DATE AND TIME

Mon., 29/04/2019, 9:30 am –

Tue., 30/04/2019, 5:00 pm AEST

Location

Saxons Training Facilities

Level 8

500 Collins Street 
Melbourne, Victoria 3000 

SYDNEY: 26–27 March

DATE & TIME

Tue., 26/03/2019, 9:30 am –

Wed., 27/03/2019, 5:00 pm AEST

LOCATION

BCA

Level 1

65 York Street

Sydney, NSW 2000

About the Course

This class will explore the many unique applications and extensions of the randomForest package, many of which are implemented in R.

Access to these methods allows the user to easily solve problems not susceptible to other methods, including deep learning.

Topics will include:

  1. A brief overview of the random forest algorithm.

  2. Out-of-sample estimates on training data, and applications in fraud, risk and outlier detection—random forests can make confident predictions on training data, unlike most other methods.

  3. Single-model quantile regression—estimating a full distribution, not just the mean. Vital for risk-based estimation.

  4. The proximity matrix—a powerful visualisation, clustering and insights tool unique to random forests.

  5. Random forests as an unsupervised learning method—outlier detection and clustering when there are no target values—vital for fraud detection.

  6. ranger, a fast, flexible implementation of random forests in R.

  7. extraTrees (Extremely Randomized Trees), an extension to random forests that often adds more accuracy.

  8. Dealing with small data sets and small classes.

A range of other topics, including recent works and extensions of existing packages, may also be covered.

Trainees are expected to be familiar with R, the basics of machine learning and out-of-sample error estimation, and the basic workings of the random forest algorithm.

Testimonials

Having studied stats at Uni I was surprised how far the field has progressed in the last few years, particularly in the area of big data. The great thing about Eugene’s course is I left with a sense that I was up to date with the latest big data modelling concepts but more importantly could also deploy them with some confidence using R. Eugene also made it clear he was available to answer questions after the course, so you are not left hanging. I would absolutely recommend this!
—Damon Rasheed, CEO, Rate Detective

For someone who does not come from an IT background R is a terrifying program. Before doing the Introduction to R course I had previously done other courses in R but always found myself in over my head because they assumed a high level of program experience (even course that required no prior programming knowledge). This course is not like that at all. It starts at ground zero and teaches you everything you need to know to be able to use R confidently in your everyday workplace. It is a must attend for anyone who wants use R!
—Alix Duncan

Data science can be a challenging topic but Eugene’s “Introduction to Machine Learning” course turns complex statistical models into plain English. The course contents and presentation were accessible and I enjoyed the mixture of hands-on rattle() exercises, the challenge of building multiple models with real life data, and the salient theory whiteboard discussions created many “aha” moments.

It was a great introductory course and it gave me with a better grasp of Machine Learning in general, a great framework for thinking about it and practical hands-on skills that I can put to immediate use. I wish I had done this course sooner.

—Charl Swart, Director of Business Operations, Unisys Credit Services

Course instructor

Courses are taught by Dr Eugene Dubossarsky and his hand-picked team of highly skilled instructors.

About our training

Eugene Dubossarsky’s courses are unlike those offered in universities, online, or by private providers. His data-science classes, in particular, give clients not just knowledge of a process, but the real power of understanding the underlying concepts, allowing them to confidently practice, manage, promote and risk-assess data science.

Dr Dubossarsky says “the way many courses teach data science is like teaching people to memorise and recite poetry in a language they do not understand”. By contrast, he confers an understanding of that language, taught in an intuitive, accessible way that leaves trainees with an instinct for data science. Keeping formulae and mathematics to a bare minimum and taking an intuitive, visual approach, Eugene’s courses deliver a compressed mentoring experience as much as they do content. This is difficult for an average trainer to replicate. Trainees benefit from his extensive knowledge and over 20 years of commercial data-science experience, as well as his unique teaching style.

The resulting testimonials speak for themselves, and candidates come from all walks of life: CEOs, general managers, salespeople, IT professionals, marketing staff, public servants and of course people from many functions in the finance world. These testimonials are extensive, and many more are available on request. With specific regard to finance, Eugene has mentored and advised senior leaders and their teams in a number of major Australian banks.

Questions and further details

Meals and refreshments

Catered morning tea and lunch are provided on both days of the course. Please notify us at least a week ahead if you have any special dietary requirements.

Variation

Course material may vary from advertised due to demands and learning pace of attendees. Additional material may be presented, along with or in place of advertised.

Frequently asked question(s) (FAQ(s))

Do I need to bring my own computer?

There’s no need to bring your own laptop or PC. Our courses take place in modern, professional training facilities that have all the computing equipment you’ll need.

You can schedule a call with a Black Cat Data Career Consultant to discuss this course.

Enrol Now