Library Load

library(tidytext)
## Warning: package 'tidytext' was built under R version 4.6.1
library(tidyverse)
## Warning: package 'tidyverse' was built under R version 4.6.1
## Warning: package 'ggplot2' was built under R version 4.6.1
## Warning: package 'tibble' was built under R version 4.6.1
## Warning: package 'tidyr' was built under R version 4.6.1
## Warning: package 'purrr' was built under R version 4.6.1
## Warning: package 'dplyr' was built under R version 4.6.1
## Warning: package 'forcats' was built under R version 4.6.1
## Warning: package 'lubridate' was built under R version 4.6.1
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.2.1     ✔ readr     2.2.0
## ✔ forcats   1.0.1     ✔ stringr   1.6.0
## ✔ ggplot2   4.0.3     ✔ tibble    3.3.1
## ✔ lubridate 1.9.5     ✔ tidyr     1.3.2
## ✔ purrr     1.2.2     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(widyr)
## Warning: package 'widyr' was built under R version 4.6.1
library(igraph)
## Warning: package 'igraph' was built under R version 4.6.1
## 
## Attaching package: 'igraph'
## 
## The following objects are masked from 'package:lubridate':
## 
##     %--%, union
## 
## The following objects are masked from 'package:dplyr':
## 
##     as_data_frame, groups, union
## 
## The following objects are masked from 'package:purrr':
## 
##     compose, simplify
## 
## The following object is masked from 'package:tidyr':
## 
##     crossing
## 
## The following object is masked from 'package:tibble':
## 
##     as_data_frame
## 
## The following objects are masked from 'package:stats':
## 
##     decompose, spectrum
## 
## The following object is masked from 'package:base':
## 
##     union
library(ggraph)
## Warning: package 'ggraph' was built under R version 4.6.1
library(checkdown)
## Warning: package 'checkdown' was built under R version 4.6.1
library(dplyr)
library(purrr)
library(tidyr)
library(lubridate)
library(tibble)
library(base)

library(tuber)
library(readr)
library(ggplot2)
library(wordcloud)
## Loading required package: RColorBrewer
library(RColorBrewer)
library(tm)
## Loading required package: NLP
## 
## Attaching package: 'NLP'
## 
## The following object is masked from 'package:ggplot2':
## 
##     annotate
library(httpuv)
library(stringr)

Part 1: Reading Analysis (https://www.r-bloggers.com/2021/02/text-analysis-with-r/)

In this article, I learned the importance of automating text analyses. Because it is a tedious process to do by hand, or say, manually, using R to complete the task of analyzing text data made sense. In this article, he used OnePlus phones as a case study because the author is/was a customer of theirs. This was particularly interesting to me because I’d want to see what others are saying about a brand that I purchase from and doing it in this way could have meaningful results.

Because words like ‘the’, ‘and’, ‘a’, ‘to’, don’t necessarily provide sentiment, the author filtered them out for easier analysis and visualization. Overall, the article shows that learning text analysis with R can help business analysts extract meaningful insights from unstructured text data and improve decision-making through automation.

Part 2: Word Co-Occurrence Analysis

Load Comments

comments_clean <- read_csv("comments_clean.csv")
## Rows: 1377 Columns: 5
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr  (3): authorDisplayName, textOriginal, id
## dbl  (1): likeCount
## dttm (1): publishedAt
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Step 1: Tokenize

custom_stop_words <- tibble(
  word = c("video", "like")
)

comments_words <- comments_clean %>%
  select(id, textOriginal) %>%
  unnest_tokens(word, textOriginal) %>%
  anti_join(stop_words, by = "word") %>%
  anti_join(custom_stop_words, by = "word") %>%
  filter(!str_detect(word, "^[0-9]+$"))

comments_words %>%
  count(word, sort = TRUE) %>%
  head(20)
## # A tibble: 20 × 2
##    word        n
##    <chr>   <int>
##  1 band      125
##  2 music     118
##  3 scott     112
##  4 boy       110
##  5 fall      107
##  6 ian        90
##  7 song       89
##  8 love       86
##  9 time       85
## 10 rock       83
## 11 andy       79
## 12 awesome    79
## 13 anthrax    70
## 14 fob        70
## 15 lol        70
## 16 metal      67
## 17 fucking    62
## 18 joe        61
## 19 guys       56
## 20 album      52

Step 2: Word Pairs

comments_pairs <- comments_words %>%
  pairwise_count(word, id, sort = TRUE, upper = FALSE)

comments_pairs %>%
  head(20)
## # A tibble: 20 × 3
##    item1   item2       n
##    <chr>   <chr>   <dbl>
##  1 fall    boy        90
##  2 scott   ian        77
##  3 time    die        39
##  4 joe     andy       28
##  5 keith   buckley    25
##  6 brian   posehn     23
##  7 jon     lajoie     22
##  8 boy     anthrax    22
##  9 anthrax die        22
## 10 time    anthrax    21
## 11 fall    band       20
## 12 boy     band       20
## 13 fall    anthrax    19
## 14 anthrax scott      18
## 15 boy     die        17
## 16 andy    hair       17
## 17 music   band       16
## 18 band    fob        16
## 19 music   boy        15
## 20 time    boy        15

Step 3: Network Graph

comments_pairs %>%
  filter(n > 5) %>%
  graph_from_data_frame() %>%
  ggraph(layout = "fr") +
  geom_edge_link(aes(edge_alpha = n, edge_width = n),
                 show.legend = FALSE) +
  geom_node_point() +
  geom_node_text(aes(label = name), repel = TRUE) +
  labs(title = "Word Co-occurrence Network: YouTube Comments")
## Warning: The `trans` argument of `continuous_scale()` is deprecated as of ggplot2 3.5.0.
## ℹ Please use the `transform` argument instead.
## This warning is displayed once per session.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

Reflection Questions

  1. The additional words that I added manually were “video” and “like.” It made sense to remove these words because the audience is aware that it is a YouTube video in which the analysis takes place, and depending on region, “like” can be used as a filler word instead of something used to make a comparison or to declare whether or not they enjoyed the video.

  2. I changed the filter value from n > 1 to n > 5. This means the graph only shows word pairs that appeared together more than five times. The result looked better because it removed weaker connections and made the network more evenly spaced and easier to interpret. My final value was 5 because it reduced clutter while still leaving enough word pairs to analyze.