Overview
This two-day course is aimed at not only teaching an understanding of some of the most common machine learning techniques, but also the approach to implementing machine learning. During this course, attendees will learn how to define a problem and prepare data, the range of techniques available for solving common problems and the approaches to take to evaluate models and achieve the best results possible.
Machine Learning can be applied to data in a whole range of fields from Finance to Pharmaceutical, Retail to Marketing, Sports to Travel and many, many more! This course is aimed at anyone interested in applying machine learning methods to their data in order to: gain deeper insight, make better decisions or build data products.
This workshop is delivered by our training partner Jumping Rivers
Outline
Introduction to analytics
A general introduction into analytics and some of the techniques that are in common use.
Single regression problems
Simple and multiple linear regression and model diagnostics.
Model selection and assessment
Cross validation and bootstrapping. Penalised regression and shrinkage.
Classification
KNN, clustering, logistic regression, Linear Discriminant analysis and associated diagnostics.
Advanced regression techniques
Polynomial regression, splines, local regression, GAMs, trees and random forests.
Workflow development
Throughout we will develop a workflow of training, testing and assessing models that can be extended to techniques not directly covered.
Course structure
Day 1
- Introduction to Analytics
- Simple regression problems
- Model selection and assessment
Day 2
- Model selection and assessment
- Classification
- Advanced regression techniques
Requirements
It will be assumed that participants are familiar with R. For example, inputting data, basic visualisation, basic data structures and use of functions. Attending the Introduction to R course will provide a sufficient background, but the programming with R will be helpful.
