This lesson is being piloted (Beta version)

Machine Learning with R

Instructor

  • Instructor: Tue Vu, PhD
  • Office: 119 Ford Hall, SMU
  • Email: tuev@smu.edu

Workshop Description

This workshop presents the overall workflow of Data Science and Machine Learning in R programming language from data collection, data cleansing, data wrangling, data partition to Supervised/Unsupervised Machine Learning: pre/postprocessing and present the final output via R-Markdown, Github markdown or HTML/PDF format. The workflow can be done via RStudio on students’ PC or SMU HPC Open OnDemand running on ManeFrame 3 platform to utilize the computational power of M3 HPC. By the end of this workshop, students are given a chance to work on real projects using Kaggle’s datasets to practice using the workflow from A-Z

Pre-requisite for the course is “Introduction to R programming and Visualization”, recorded as access via SMU panopto:

Course Outline

Topic

Description
Setup Preparing for the course
1. Introduction to Data Science workflow with R What is the overall workflow of Data Science
2. Introduction to Caret What is Caret
3. Data Partition with caret What is Data Partition
4. Evaluation Metrics with caret How do we measure the accuracy of ML model
5. Supervised Learning with Continuous Output How to train a Machine Learning model using Regression method
6. Supervised Learning with Categorical Output How to train a Machine Learning model with categorical output
7. Unsupervised Learning What is Unsupervised Learning in Machine Learning model
8. Regularization and Variable Selection Why do we need Regularization and Variable Selection in ML model
9. Kaggle online competition: Supervised Learning How to participate in a Kaggle online compeition
10. Kaggle online competition: Unsupervised Learning How to participate in a Kaggle online compeition
11. Fundamental Text Mining using R What is Text Mining and how to use R to work with that
12. Part of Speech Tagging How to Perform POS using R
Finish

The actual schedule may vary slightly depending on the topics and exercises chosen by the instructor.