Some weeks ago I had a presentation at my work place about “R for data science” that I’d like to share with you. I’ve written the slides in R and rmarkdown and uploaded them to rpubs.com. I chose to use rmarkdown for my slides although we have great company PowerPoint templates, because I wanted to show how you can integrate creating slides in your data analysis workflow. In this post I share my presentation, some background information and how I prepared for the talk with you!
Writing the presentation in R + rmarkdown
Since I have used markdown and rmarkdown before, preparing the slides was pretty straight forward. I only had an issue with image sizes which I could resolve with an answer from stackoverflow. Stackoverflow answer: Rpresentation in Rstudio – Make image fill out the whole screen.
I used all the standard settings but for the next presentation I would definitely change the font size. We were sitting in a large meeting room and I think the people in the back could hardly read the text on my slides.
Practicing the presentation
Due to the fact that I am always super nervous when it comes to presenting in front of people (that’s not an exaggeration) I wanted to practice my presentation first. My cats have listened to some of my presentations before but I was not really sure they could ask the right questions. So I thought we have this great Data Science Learning Club, which should help learn people data science things they are not yet good or comfortable at. For me, presenting is part of data science so I asked in our Learning Club Slack Channel if anyone was interested in listening to my presentation. Two days before the actual presentation I was able to practice my presentation in front of a (very small) machine learning audience. This helped me a lot 🙂
So if you have the same problem, just find someone online where you can practice your presentation using Skype, Hangout or whatever!
Every year we can some work time on further education like university or online courses. The idea of this presentation was to share what we have learned in these courses so that everyone gets an overview. Since the regression courses were mostly theoretical, I thought an introduction to R might be a benefit for more people in our department. So please don’t be surprised why there is a slide with three regression courses at the beginning of the presentation 🙂
The focus of the presentation is:
- Data handling with dplyr
- Data visualisation with ggplot2
- Model training and evaluation with caret
- Presentation of results
My slides are located here.
At the end of the presentation I shared some links which I think are a good starting point if you want to get to know R!
- R 4 Data Science (book)
- ggplot2 extensions (you might get a security warning)
- tidyverse (release post)
- the caret package
- Markdown basics
- RStudio cheatsheets (awesome!)