Chapter 5 Performance Tier and Gems

Experts [ 28% ] followed by Masters [ 25% ] contribute most of the Hidden Gems.Grand Masters [ 18% ] and Contributors [ 18% ] follow next. The Novice Tier and the Kaggle Team has also contributed to the Gems

TotalNoOfRows <- nrow(gems_users)

PerformanceTier <- c(NA,0,1,2,3,4,5)
PerformanceTier_Name <- c("NoTier","Novice","Contributor","Expert","Master","GrandMaster","KaggleTeam")

df_Performance_Tier <-data.frame(PerformanceTier, PerformanceTier_Name)

gems_user_performance <- inner_join(gems_users,df_Performance_Tier)


p1 <- gems_user_performance %>%
  group_by(PerformanceTier_Name) %>%
  summarise(Percentage = n()/TotalNoOfRows *100) %>%
  arrange(desc(Percentage)) %>%
  ungroup() %>%
  mutate(PerformanceTier_Name = reorder(PerformanceTier_Name,Percentage))

p1 %>%
  filter(!is.na(PerformanceTier_Name)) %>%
  ggplot(aes(x = PerformanceTier_Name,y = Percentage, fill = (PerformanceTier_Name) )) +
  geom_bar(stat='identity',colour="white")  +
  geom_label(aes(label = paste0("( ",round(Percentage,2)," %)",sep=""))) +
  labs(x = 'Performance Tier', 
       y = 'Percentage', 
       title = 'Performance Tier and Percentage') +
  coord_flip() + 
  theme_fivethirtyeight(base_size = 15) +
  theme(legend.position = "none")