knitr::opts_chunk$set(echo = TRUE)
The results appear accurate, and match the vignette.
To a novice, it is not entirely obvious how the results are computed as the topic and terminology are complex (example: "The tidyDTM function allows the user to construct a dtm for a given sparsity."), one already has to be familiar with the textmining terminology, which they probably are if they downloaded the package. But the vignette is detailed and clear on each function.
After reading the instructions and using devtools::install_github("williamcsevier/textclass", dependencies = TRUE, build_vignettes = TRUE)
the analytic worked perfectly.
No.
If I was to get a little more experienced on topic-modeling and text-modeling it seems like this would be a pretty cool tool. The app on shinyapps is also cool.
Yes the visualizations and plots are very good.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.