From Magic to Mixed Feelings: Analyzing ‘One Hundred Years of Solitude’ Reviews

How readers experience the novel: A deep dive into emotional responses, writing complexity, and thematic connections across different rating categories.

A comprehensive analysis of reader reviews for ‘One Hundred Years of Solitude’, examining emotional patterns, writing complexity, and common themes through data visualization of Goodreads and LibraryThing reviews.

Figure 1: Data visualization analyzing reviews of One Hundred Years of Solitude with four plots: (1) Distribution of emotional content by rating category, showing positive emotions dominating higher ratings; (2) Emotional flow through reviews, illustrating a mix of joy, trust, and sadness across the text; (3) Review complexity by rating, indicating longer sentences in positive reviews; (4) Common word pairs in reviews, highlighting frequent terms such as ‘family’, ‘buendía’, and ‘realism’.

Update: This post has been updated based on feedback from the #SWDchallenge community. The changes include: - Fixed the chart legends that were inadvertently left out during one iteration.

Steps to Create this Graphic

1. Load Packages & Setup

if (!require("pacman")) install.packages("pacman")
pacman::p_load(
  tidyverse,         # Easily Install and Load the 'Tidyverse'
  ggtext,            # Improved Text Rendering Support for 'ggplot2'
  showtext,          # Using Fonts More Easily in R Graphs
  scales,            # Scale Functions for Visualization
  glue,              # Interpreted String Literals
  here,              # A Simpler Way to Find Your Files
  janitor,           # Simple Tools for Examining and Cleaning Dirty Data
  skimr,             # Compact and Flexible Summaries of Data
  camcorder,         # Record Your Plot History
  textcat,           # N-Gram Based Text Categorization
  ggdist,            # Visualizations of Distributions and Uncertainty # Visualizations of Distributions and Uncertainty # Visualizations of Distributions and Uncertainty
  tidytext,          # Text Mining using 'dplyr', 'ggplot2', and Other Tidy Tools # Text Mining using 'dplyr', 'ggplot2', and Other Tidy Tools # Text Mining using 'dplyr', 'ggplot2', and Other Tidy Tools
  patchwork          # The Composer of Plots # The Composer of Plots # The Composer of Plots
) 

# Source utility functions
suppressMessages(source(here::here("R/utils/fonts.R")))
source(here::here("R/utils/social_icons.R"))
source(here::here("R/utils/image_utils.R"))
source(here::here("R/themes/base_theme.R"))

2. Read in the Data

goodreads <- read_csv(
  here::here("data/goodreads_reviews_full.csv"))
  
librarything <- read_csv(
  here::here("data/librarything_reviews_full.csv"))

# Combine the datasets
combined_reviews <- bind_rows(goodreads, librarything)

3. Examine the Data

glimpse(goodreads)
glimpse(librarything)
glimpse(combined_reviews)

4. Tidy Data

combined_reviews_clean <- combined_reviews |>
  # Combine 'star_rating' and 'numeric_rating' into a single 'rating' column
  mutate(rating = coalesce(star_rating, numeric_rating)) |>
  # Convert 'review_date' to Date format
  mutate(review_date = lubridate::mdy(review_date)) |>
  # Standardize column names
  rename(
    reviewer = reviewer_name,
    date = review_date,
    text = review_text
  ) |>
  # Clean up review text  
  mutate(
    text = str_squish(text), # Remove extra whitespace
    text = tolower(text),    # Convert to lowercase
    text = str_replace_all(text, "[^a-zA-Z0-9 .,!?']", "") # Remove special characters
  ) |>
  # Select and reorder columns
  select(reviewer, date, rating, text, source) |>
  # Remove duplicate rows
  distinct() |> 
  mutate(
    language = textcat(text),             # Add detected language as a new column
    word_count = str_count(text, "\\S+")  # Count words in text
    ) |>  
  filter(language == "english")           # Keep only English reviews


# Housekeeping
rm(goodreads, librarything, combined_reviews)


# Prepare text data for sentiment analysis
review_sentiments <- combined_reviews_clean |>
  unnest_tokens(word, text) |>
  anti_join(stop_words) |>
  inner_join(get_sentiments("nrc")) |>
  # Add rating categories for comparison
  mutate(rating_category = case_when(
    rating <= 2 ~ "Negative (1-2)",
    rating == 3 ~ "Neutral (3)",
    rating >= 4 ~ "Positive (4-5)"
  ))

# 1. Revised Complexity Analysis
complexity_analysis <- combined_reviews_clean |>    
  mutate(
    sentences = str_count(text, "[.!?]+"),
    words_per_sentence = word_count / sentences,
    rating_category = factor(case_when(
      rating <= 2 ~ "Negative (1-2)",
      rating == 3 ~ "Neutral (3)",
      rating >= 4 ~ "Positive (4-5)"
    ), levels = c("Negative (1-2)", "Neutral (3)", "Positive (4-5)"))
  ) |>
  filter(is.finite(words_per_sentence))

# 2. Sentiment Flow (keeping existing structure, updating colors)
sentiment_flow <- review_sentiments |>
  mutate(
    theme = case_when(
      sentiment %in% c("joy", "trust", "anticipation") ~ "positive",
      sentiment %in% c("anger", "fear", "disgust") ~ "negative",
      TRUE ~ "neutral"
    )
  ) |>
  count(rating_category, theme) |>
  group_by(rating_category) |>
  mutate(prop = n/sum(n)) |>
  ungroup()

# 3. Temporal Pattern (keeping existing structure)
temporal_pattern <- review_sentiments |>
  group_by(reviewer) |>
  mutate(
    position = row_number(),
    position_pct = position/n()
  ) |>
  count(position_pct = round(position_pct, 2), sentiment) |> 
  ungroup()

# 4. Simplified Bigram Network
bigram_graph <- combined_reviews_clean |>
  unnest_tokens(bigram, text, token = "ngrams", n = 2) |>
  separate(bigram, c("word1", "word2"), sep = " ") |>
  filter(
    !word1 %in% stop_words$word,
    !word2 %in% stop_words$word,
    !is.na(word1),
    !is.na(word2)
  ) |>
  count(word1, word2, sort = TRUE) |>
  filter(n >= 4) |>  # Increased threshold
  slice_head(n = 15)  # Take only top 15 pairs

5. Visualization Parameters

### |-  plot aesthetics ----
# Get base colors with custom palette
colors <- get_theme_colors(palette = c("negative" = "#E69B95", "neutral"  = "#709BB0", "positive" = "#86B8B1"))

### |-  titles and caption ----
title_text   <- str_glue("From Magic to Mixed Feelings: Analyzing 'One Hundred Years of Solitude' Reviews") 

subtitle_text <- str_glue(
  "How readers experience the novel: A deep dive into emotional responses, writing complexity, and thematic\n
connections across different rating categories",
  
  "\n\n**Note**: This analysis is based on a small sample of 42 reviews, collected from Goodreads and LibraryThing\n
as of January 3, 2025.")

# Create caption
caption_text <- create_swd_caption(
  year = 2025,
  month = "Jan",
  source_text = "Source: Scrapped from goodreads & librarthing"
)


# |- fonts ----
setup_fonts()
fonts <- get_font_families()

### |-  plot theme ----
# Start with base theme
base_theme <- create_base_theme(colors)
            

# Add weekly-specific theme elements
weekly_theme <- extend_weekly_theme(
    base_theme,
    theme(
      plot.margin         = margin(t = 10, r = 20, b = 10, l = 20),
      axis.title.x        = element_text(margin = margin(10, 0, 0, 0), size = rel(1.1), 
                                         color = colors$text, family = fonts$text, face = "bold", hjust = 0.5),
      axis.title.y        = element_text(margin = margin(0, 10, 0, 0), size = rel(1.1), 
                                         color = colors$text, family = fonts$text, face = "bold", hjust = 0.5),
      axis.text           = element_text(size = rel(0.8), color = colors$text),
      axis.line.x         = element_line(color = "#252525", linewidth = .3),
      axis.ticks.x        = element_line(color = colors$text),  
      axis.title          = element_text(face = "bold"),
      panel.grid.minor    = element_blank(),
      panel.grid.major    = element_blank(),
      panel.grid.major.y  = element_line(color = "grey85", linewidth = .4)
      )
)
      

# Set theme
theme_set(weekly_theme)

6. Plot

# 1. Sentiment Flow Plot
p1 <- sentiment_flow |>  
  ggplot(aes(x = rating_category, y = prop, fill = theme)) +
  geom_col(position = "fill", alpha = 0.9) +
  scale_fill_manual(values = colors$palette) +
  scale_y_continuous(labels = scales::percent) +
  labs(
    title = "<b>Distribution of Emotional Content by Rating</b>",
    fill = "Emotional Theme",
    x = "Rating Category",
    y = "Proportion of Emotions",
  ) +
  theme_minimal() +
  theme(
    plot.title = element_markdown(size = rel(1)),
    legend.position = "right",
    plot.margin = margin(t = 10, r = 10, b = 20, l = 10)
  )

# 2. Temporal Pattern Plot
p2 <- temporal_pattern |> 
  ggplot(aes(x = position_pct, y = n, fill = sentiment)) +
  geom_area(position = "fill", alpha = 0.7) +
  scale_fill_brewer(palette = "RdYlBu") +
  scale_x_continuous(
    labels = scales::percent,
    breaks = c(0, 0.25, 0.5, 0.75, 1),
    expand = c(0, 0)
  ) +
  scale_y_continuous(
    labels = scales::percent,
    expand = c(0, 0)
  ) +
  labs(
    title = "<b>Emotional Flow Through Reviews</b>",
    x = "Relative Position in Review",
    y = "Proportion of Emotions",
    fill = "Emotion"
  ) +
  theme_minimal() +
  theme(
   plot.title = element_markdown(size = rel(1)),
    legend.position = "right",
    panel.grid.minor = element_blank(),
    plot.margin = margin(t = 10, r = 10, b = 20, l = 10)
  )

# 3. Complexity Analysis Plot
p3 <- complexity_analysis |> 
  ggplot(aes(x = words_per_sentence, y = rating_category, fill = rating_category)) +
  stat_gradientinterval(
    aes(color = after_scale(fill)), 
    point_size = 1.2,
    alpha = 0.3,
    point_alpha = 0.7
  ) +
  scale_fill_manual(
    values = c(
      "Negative (1-2)" = "#E69B95",
      "Neutral (3)"    = "#709BB0",
      "Positive (4-5)" = "#86B8B1"
    )
  ) +
  labs(
    title = "<b>Review Complexity by Rating</b>",
    x = "Words per Sentence",
    y = NULL,
    fill = "Rating Category"
  ) +
  theme_minimal() +
  theme(
    plot.title = element_markdown(size = rel(1)),
    legend.position = "right",
    panel.grid.major.y = element_blank(),
    plot.margin = margin(t = 10, r = 10, b = 20, l = 10)
  )

# 4. Bigram Network Plot
p4 <- bigram_graph |> 
  ggplot(aes(x = word1, y = word2)) +
  geom_point(aes(size = n), color = colors$palette["neutral"], alpha = 0.7) +
  scale_size_continuous(range = c(2, 6)) +
  labs(
    title = "<b>Common Word Pairs in Reviews</b>",
    size = "Frequency"
  ) +
  theme_minimal() +
  theme(
    plot.title = element_markdown(size = rel(1)),
    legend.position = "right",
    axis.text.x = element_text(hjust = 1),
    panel.grid = element_line(color = "grey90"),
    plot.margin = margin(t = 10, r = 10, b = 20, l = 10)
  )

# Combine plots 
combined_plots <- (p1 + p2) /
  (p3 + p4) +
  plot_annotation(
    title = title_text,
    subtitle = subtitle_text,
    caption = caption_text,
    theme = theme(
      plot.title = element_markdown(
        family = "title",
        face = "bold", 
        size = rel(1.7),
        color = colors$title,
        margin = margin(b = 10)
      ),
      plot.subtitle = element_markdown(
        family = "subtitle",
        size = rel(1.1),
        color = colors$subtitle, 
        margin = margin(b = 20),
        lineheight = 1.1
      ),
      plot.caption = element_markdown(
        family = "caption",
        size = 10, 
        color = colors$caption,
        margin = margin(t = 20),
        hjust = 0.5,
        lineheight = 1.2
      )
    )
  ) &
  theme(plot.background = element_rect(fill =colors$background, color = NA))

7. Save

### |-  plot image ----  

source(here::here("R/image_utils.R"))
save_plot_patchwork(combined_plots, type = 'swd', year = 2025, month = 01, 
                    width = 12, height = 12)

8. Session Info

TipExpand for Session Info

R version 4.4.1 (2024-06-14 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 11 x64 (build 22631)

Matrix products: default


locale:
[1] LC_COLLATE=English_United States.utf8 
[2] LC_CTYPE=English_United States.utf8   
[3] LC_MONETARY=English_United States.utf8
[4] LC_NUMERIC=C                          
[5] LC_TIME=English_United States.utf8    

time zone: America/La_Paz
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices datasets  utils     methods   base     

other attached packages:
 [1] patchwork_1.3.0 tidytext_0.4.2  ggdist_3.3.2    textcat_1.0-9  
 [5] camcorder_0.1.0 skimr_2.1.5     janitor_2.2.0   here_1.0.1     
 [9] glue_1.8.0      scales_1.3.0    showtext_0.9-7  showtextdb_3.0 
[13] sysfonts_0.8.9  ggtext_0.1.2    lubridate_1.9.3 forcats_1.0.0  
[17] stringr_1.5.1   dplyr_1.1.4     purrr_1.0.2     readr_2.1.5    
[21] tidyr_1.3.1     tibble_3.2.1    ggplot2_3.5.1   tidyverse_2.0.0
[25] pacman_0.5.1   

loaded via a namespace (and not attached):
 [1] tidyselect_1.2.1     farver_2.1.2         fastmap_1.2.0       
 [4] janeaustenr_1.0.0    digest_0.6.37        timechange_0.3.0    
 [7] lifecycle_1.0.4      rsvg_2.6.1           tokenizers_0.3.0    
[10] magrittr_2.0.3       compiler_4.4.0       rlang_1.1.4         
[13] tools_4.4.0          utf8_1.2.4           yaml_2.3.10         
[16] knitr_1.49           labeling_0.4.3       htmlwidgets_1.6.4   
[19] curl_6.0.0           bit_4.5.0            RColorBrewer_1.1-3  
[22] xml2_1.3.6           repr_1.1.7           withr_3.0.2         
[25] grid_4.4.0           fansi_1.0.6          colorspace_2.1-1    
[28] cli_3.6.3            rmarkdown_2.29       crayon_1.5.3        
[31] generics_0.1.3       rstudioapi_0.17.1    textdata_0.4.5      
[34] tzdb_0.4.0           commonmark_1.9.2     parallel_4.4.0      
[37] ggplotify_0.1.2      yulab.utils_0.1.8    base64enc_0.1-3     
[40] vctrs_0.6.5          Matrix_1.7-0         jsonlite_1.8.9      
[43] slam_0.1-55          gridGraphics_0.5-1   hms_1.1.3           
[46] bit64_4.5.2          systemfonts_1.1.0    magick_2.8.5        
[49] gifski_1.32.0-1      codetools_0.2-20     distributional_0.5.0
[52] stringi_1.8.4        gtable_0.3.6         munsell_0.5.1       
[55] pillar_1.9.0         rappdirs_0.3.3       htmltools_0.5.8.1   
[58] R6_2.5.1             rprojroot_2.0.4      vroom_1.6.5         
[61] evaluate_1.0.1       lattice_0.22-6       markdown_1.13       
[64] SnowballC_0.7.1      gridtext_0.1.5       tau_0.0-26          
[67] snakecase_0.11.1     renv_1.0.3           Rcpp_1.0.13-1       
[70] svglite_2.1.3        xfun_0.49            fs_1.6.5            
[73] pkgconfig_2.0.3     

9. GitHub Repository

TipExpand for GitHub Repo

The complete code for this analysis is available in swd_2025_01.qmd. For the full repository, click here.

10. References

TipExpand for References

The web scraping scripts used to collect the review data: - Goodreads: goodreads_web_scraping.R - LibraryThing: librarything_web_scraping.R

Data Sources: - Goodreads: One Hundred Years of Solitude Reviews - LibraryThing: One Hundred Years of Solitude Reviews