Kaggle Hidden Gems

library(tidyverse)

hidden_gems <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2022/2022-04-26/hidden_gems.csv')

df <- hidden_gems%>%
  count(vol,date,title,review,author_kaggle,author_name)

df1 <- df %>%  #pull(date)%>%summary
  mutate(ym=zoo::as.yearmon(date),
         year=lubridate::year(date))%>%
  unnest_tokens(word,review) %>%
  anti_join(get_stopwords()) %>%
  filter(!str_detect(word,"kaggle|c|https|2020|2021|also|end|r")) %>%
  count(author_name,word,sort=T) %>%

 inner_join(bing) %>%
  mutate(sentiment2=sentiment)%>%
  spread(sentiment, n, fill = 0) %>%
  mutate(sentiment = positive - negative) %>%
  mutate(index=row_number())%>%
  relocate(index) %>% #count(word,sort=T)
  mutate(word=reorder(word,-index)) 

my_words<-df1 %>%
  count(word,sort=T)%>%
  filter(n>1) %>%
  select(-n) %>%
  unlist()
  
  
df2 <- df1 %>%
   filter(word %in% my_words) 

Notches are used to compare groups; if the notches of two boxes do not overlap, this suggests that the medians are significantly different.

df2 %>%
  mutate(word=toupper(word))%>%
  mutate(word=as.factor(word),
         word=reorder(word,-index))%>% #count(index,sort=T)
  ggplot(aes(x=index,y=fct_reorder(word,sentiment),fill=sentiment2))+ # 
  geom_boxplot(size=0.5,
               outlier.size = 0.3,outlier.shape = 21,
               notch = T) +
  geom_jitter(size=0.3,shape=21,stroke=0.5)+
  coord_cartesian(xlim = c(-1,260),ylim=c(0,35),clip="off")+
  labs(title="How vary are words in the Hidden Gems Reviews?",
       subtitle="Some words repeated themselves more frequently than others,\nwith significantly different medians.",
       caption="#30DayChartChallenge 2022 #Day28 - Deviations\nDataSource: #TidyTuesday Week17 - Hidden Gems\nDataViz: Federica Gazzelloni (@fgazzelloni)",
       x="",y="",fill="Sentiment")+
  tvthemes::scale_fill_rickAndMorty()+
  tvthemes::theme_rickAndMorty()+
  theme(plot.background = element_rect(fill="grey80"),
        legend.background = element_rect(fill=NA),
        legend.position = c(0.9,0.9),
        plot.title = element_text(face="bold",size=22),
        plot.title.position = "plot",
        plot.subtitle = element_text(size=12),
        axis.text.x = element_blank())+
  annotate("text",x=-70,y=-2.5,label="How to read this graph:\n- the notches represent the median of the frequency of words\nfound in the Hidden Gems reviews by Author\n- the sidebars represent their deviations",
           hjust=0,size=3)

ggsave("day28_deviations.png",
       dpi=320,
       width = 8,
       height = 10)