After the Leaflet example I was given another assignment in the Coursera Developing Data Products course. This time the goal was to use Shiny to demonstrate some interactive data analysis and visualisation.
I picked two areas I’m fond of, books and sentiment analysis, and combined them to produce the Gutenberg Sentiment Analyser!
Here is how to use it:
- Search for an author and pick one.
- Similarly in the word cloud tab the books are dissected and visualised.
- In the sentiment tab an analysis is generated and plotted for all the author’s books from beginning to end of each with normalised book length.
It makes use of the following R packages:
- gutenbergr
- tidytext
- wordcloud
Some examples 🔗
Here’s a word cloud of Jane Austen’s works:
Dostoyevsky in particular is somewhat depressing:
To be honest I didn’t have much time to do this so it’s all a bit rushed. The graphs don’t work well for you prolific authors. I would like to figure out a better way of showing sentiment.
The source code is here.