0:00 - Start
1:32 - Workshop Goals
3:50 - Introduction to Text Mining
14:18 - How to get the code for this workshop
15:07 - CODING BEGINS
15:30 - Tokenization
16:43 - unnest_tokens()
19:18 - data cleaning
21:03 - Assign line numbers
22:22 - tokenize
23:00 - stop words
25:52 - Count word frequency
26:31 - Visualize word frequency
28:14 - Your turn
29:07 - Q/A
29:57 - Sentiment Analysis
36:04 - Visualize word frequency with a bar graph. e.g. most frequent positive and negative words
36:28 - ggplot2::geom_col() to generate bar graph
38:03 - sentiment dictionaries
40:17 - visualize sentiment when using AFINN sentiment dictionary
41:43 - Q/A part 2.
Apply the lessons of _Text Mining with R_ by Silge & Robinson. First, analyze public domain novels by Jane Austen, wrangle text-data into submission, tokenize corpora, generate word clouds, and be introduced to introductory sentiment analysis.
This Rfun case study demonstrate the utility R / Tidyverse workflows. You can use the Tidyverse as a universal reproducible interface for your analysis projects.
More Rfun at https://Rfun.library.duke.edu/
Part of the DVS Workshop Series: http://library.duke.edu/data/
LINKS
- Code for this workshop: https://github.com/libjohn/workshop_textmining
Documentation: _Text mining with R: a tidy approach_ by Julia Silge & David Robinson :: https://www.tidytextmining.com/
tidytext: Text mining using tidy tools :: https://juliasilge.github.io/tidytext/