TRAINING CATEGORIES
(Click Category to List Courses)

50 - ACC- Accredited Courses


OXI 201 - Environmental Modelling: Machine Learning and Data Mining with R

Please contact us for fees

 

Course Description

This 5-day course is focussed on applied environmental modelling. Participants will learn leading modelling techniques from the huge array of possible modelling techniques that exist. Participants will thus benefit from the trainer’s considerable experience in research and modelling by jumping directly to up-to-date modelling techniques needed for today’s world.

In recent years, there has been an exponential increase in data generation. This brings new possibilities to analyse environmental systems. Applications are wide and include improved decision-making in environmental management; environmental protection policies; and urban design. Participants will get hands-on experience and be shown the best practices for understanding, visualising and modelling environmental data.

This program is organized in cooperation with Oxford Intellect. Oxford Intellect supports higher education institutions to benefit from cutting-edge knowledge developed by leading experts in the United Kingdom. Their team is drawn from the leading universities in the United Kingdom including the University of Oxford, the London School of Economics and University College London. They benefit from a wide network of professors, lecturers and researchers who are leading figures in their field. 

Course Objectives

  • Best practice for environmental systems analysis
  • Understanding the data analysis process
  • Concepts and application of data mining
  • A broad machine learning toolbox for regression and classification tasks
  • A set of best practices for regression and classification analysis
  • Familiarisation with the R programming language

Who Should Attend?

  • Professionals and academics working and researching in the field of environmental systems
  • Professionals engaged in modelling environmental systems
  • Managers and executives who interpret and use the information generated by these models
  • Academics researching and teaching in the field of environmental systems 

Course Details/Schedule

Day 1

Essentials of modelling

  • Introduction to R and RStudio
  • Overview of programming in R
  • Installing and loading packages
  • Case study 1: modelling of the Atlantic Meridional Overturning Circulation
  • Case study 2: modelling of global magnetic disturbances in near-Earth space
  • Introduction to the tidyverse package
  • Summary statistics
  • Pipelines and pre-processing

Day 2

Modelling your data (Regression Analysis)

  • Introduction to the caret package
  • Best practices for regression analysis
  • Interpretability VS Predictability (parametric and tree-based regression models)
  • How to validate your models (data splitting, cross-validation)
  • Model comparison

Day 3

Modelling your data (Classification Analysis)

  • Best practices for classification analysis
  • Interpretability VS Predictability (parametric and tree-based classification models)
  • How to validate your models (confusion matrices and ROC plots)
  • Model comparison

Day 4

Visualising your data

  • Introduction to the ggplot2 package
  • Aesthetic mappings
  • Facets

Day 5

Examples, applications and advanced concepts

  • Specific plots
  • Applied learning examples
  • Useful resources
  • Wrap-up