This report applies Sections 1.1 through 1.3 of Text Mining with R to the five customer comments provided in the assignment. The analysis converts comments into individual words, removes common stop words, and counts and visualizes word frequencies.
Initially, each comment is a single text string. The data table preserves a comment ID so each word can be traced to its original comment.
comments <- tibble(
comment_id = 1:5,
text = c(
"I love the quality of this product. The price is reasonable, and delivery was fast.",
"The product quality is excellent, but the shipping took too long.",
"Great customer service! I would definitely recommend this product to my friends.",
"The price is too expensive. However, the product looks beautiful.",
"Fast delivery and excellent service. I am very satisfied with my purchase."
)
)
knitr::kable(comments)
| comment_id | text |
|---|---|
| 1 | I love the quality of this product. The price is reasonable, and delivery was fast. |
| 2 | The product quality is excellent, but the shipping took too long. |
| 3 | Great customer service! I would definitely recommend this product to my friends. |
| 4 | The price is too expensive. However, the product looks beautiful. |
| 5 | Fast delivery and excellent service. I am very satisfied with my purchase. |
unnest_tokens() splits the comments into words, converts
them to lowercase, and removes punctuation. The resulting table has one
word per row, while retaining the comment ID.
tidy_comments <- comments %>%
unnest_tokens(word, text)
knitr::kable(head(tidy_comments, 20))
| comment_id | word |
|---|---|
| 1 | i |
| 1 | love |
| 1 | the |
| 1 | quality |
| 1 | of |
| 1 | this |
| 1 | product |
| 1 | the |
| 1 | price |
| 1 | is |
| 1 | reasonable |
| 1 | and |
| 1 | delivery |
| 1 | was |
| 1 | fast |
| 2 | the |
| 2 | product |
| 2 | quality |
| 2 | is |
| 2 | excellent |
Common words such as “the” and “and” can obscure the topics customers
discuss. anti_join() removes words included in the tidytext
stop-word list before the remaining words are counted.
clean_comments <- tidy_comments %>%
anti_join(stop_words, by = "word")
word_counts <- clean_comments %>%
count(word, sort = TRUE) %>%
arrange(desc(n), word)
knitr::kable(word_counts, col.names = c("Word", "Frequency"))
| Word | Frequency |
|---|---|
| product | 4 |
| delivery | 2 |
| excellent | 2 |
| fast | 2 |
| price | 2 |
| quality | 2 |
| service | 2 |
| beautiful | 1 |
| customer | 1 |
| expensive | 1 |
| friends | 1 |
| love | 1 |
| purchase | 1 |
| reasonable | 1 |
| recommend | 1 |
| satisfied | 1 |
| shipping | 1 |
ggplot(word_counts, aes(x = reorder(word, n), y = n)) +
geom_col(fill = "#287D8E", width = 0.75) +
coord_flip() +
scale_y_continuous(breaks = seq(0, max(word_counts$n), by = 1)) +
labs(
title = "Word Frequencies in Five Customer Comments",
subtitle = "After removing common stop words",
x = NULL,
y = "Number of occurrences"
) +
theme_minimal(base_size = 12)
“Product” was the most frequent remaining word, appearing four times, while “quality,” “price,” “delivery,” “fast,” “excellent,” and “service” each appeared twice. Reading the original comments alongside these counts reveals praise for quality and service, as well as concerns about high prices and slow shipping.
A challenge in this workflow is that common words can dominate the
results and make the main topics harder to identify. Removing stop words
with anti_join(stop_words, by = "word") makes the frequency
table easier to interpret.
These results describe only the five supplied comments. Word frequency does not determine sentiment: for example, the two comments mentioning price express different opinions, so the original wording is needed to interpret the counts.
Julia Silge and David Robinson, Text Mining with R, Sections 1.1–1.3.