Toronto Shelter

Toronto Shelter

clearning : 날짜 변수를 year, month 로 나눔

1. 년도별, occupancy

   gruop_by(년도), summarise 합계를 구해서

   geom_point 연결

2. 월별 capcity 비율 year 세분화

   shleter 이름의 중복을 제거해주고, occupancy 와 capacity 의 year 별 합계를 구하고, 나눠서 그 비율을 구함

3. Waffle Chart - 각 sector별 shelter name 의 갯수 (count) 를 구해서 나타내는 그래프

#font 

font <-"Roboto Condensed"



#data 

tuesdata <-tidytuesdayR::tt_load(2020, week = 49)

## 
##  Downloading file 1 of 1: `shelters.csv`

tuesdata$shelters -> shelters


# Year / Month 추출 

shelters %>% 
  mutate(year = year(occupancy_date), 
         month = month(occupancy_date)) %>%  
  unite("date", year:month, sep= "/", remove = F) ->shelters




shelters %>% 
  group_by(date) %>%  #연도별 
  summarise(month_occupancy = sum(occupancy)/1000) %>% 
  mutate(date= parse_date_time(date, "ym")) %>%  
  ggplot(aes(date, month_occupancy))+
  geom_point(size=2,  color = "#371206")+
  geom_smooth(se=F, color= "#F72C25", size = 1.5)+
  theme(text = element_text(family = font),
        plot.subtitle = element_markdown(color ="#292929" ),
        plot.background = element_rect("#EBEBEB"),
        panel.background = element_blank(),
        axis.title = element_text(color = "#292929", face = "bold"),
        axis.text = element_text(color = "#292929", face = "bold"),
        panel.grid = element_blank())+
  labs(x= "Date", y= "Monthly Occupancy", 
       subtitle = "**<span style='color:#F72C25;font-size:60px;'>50%</span><br>increase in monthly shelter occupancy<br>from Jan 2017 to Dec 2019.**" ) -> plot_1

#Bar Plot: 

shelters %>% 
  filter(!is.na(capacity)) %>% 
  group_by(month, year) %>% 
  summarise(n_distinct(shelter_name),  #n_distinct 중복 개수 Count
            year_occupancy = sum(occupancy), 
            year_capacity = sum(capacity),
            rate = round(year_occupancy/year_capacity,2)) %>% 
  ggplot(aes(month, rate, fill=rate))+
  geom_bar(stat = "identity")+
  ylim(-1,1)+ 
  coord_polar()+ ## pie chart 변환 
  scale_fill_gradient(low = "#ffb950", high = "#a50104") + 
  facet_wrap(~year, ncol = 3)+
  theme(text = element_text(family = font), 
        legend.position = "none",
        axis.text.y = element_blank(),
        axis.ticks = element_blank(),
        plot.background = element_rect("#EBEBEB"),
        panel.background = element_blank(),
        axis.title = element_text(color = "#292929", face = "bold"),
        axis.text = element_text(color = "#292929", face = "bold"),
        plot.subtitle = element_markdown(face = "bold", color = "#292929"),
        strip.text.x = element_text(face = "bold", color = "#292929"),
        strip.background = element_blank(),
        panel.spacing = unit(2, "lines")) -> plot_2

plot_all <- plot_1/plot_2 


plot_all + plot_annotation(title = "Toronto Shlaters", 
                           subtitle = "As of 2019 there were 62 sheleter in Toronto",
                           theme = theme(plot.title = element_text(color = "#292929",size = 22, face = "bold", family = font),
                                plot.subtitle = element_text(color = "#292929", size = 14, family = font),
                                plot.caption = element_text(color = "grey50", face = "bold.italic", family = font),
                                plot.background = element_rect("#EBEBEB"),
                                panel.background = element_rect("#EBEBEB")))

Sample-Superstore

setwd("C:/Users/Administrator/Desktop/R Analysis")

read.csv("Sample-Superstore-Subset-Excel.csv", header = T) -> product

font <-"Roboto Condensed"

# Date 추출     

substr(product$Order.Date, 7,10) -> product$year
substr(product$Order.Date, 4, 5) -> product$month


#as.Date(product$Order.Date,  format = "%m/%d/%Y")


product %>%  
  mutate(year = as.factor(year), 
         month = as.factor(month)) %>% 
  unite("Date", year:month, sep = "/", remove = F) -> sales


my_colors <- c("#3e6487", "#829cb2", "#c7cdd1", "#edad88", "#e36c33", "#EBEBEB")

sales %>%  
 group_by(Customer.Segment, Order.Priority) %>% 
  count() %>% 
  ungroup() %>% 
  group_by(Customer.Segment) %>% 
  mutate(pct = 100*n/sum(n)) %>% 
  mutate(pct = round(pct,2)) %>% 
  mutate(bar_text = paste0(pct, "%")) %>% 
  ungroup() -> df_for_ploting


df_for_ploting %>%  
  ggplot(aes(Customer.Segment, pct, fill= Order.Priority))+
  geom_col()+
  coord_flip()+
  theme(legend.position = "top")+
  scale_fill_manual(values = my_colors[6:1], name = "")+
  theme(text = element_text(family = font))  +
  guides(fill = guide_legend(reverse = TRUE)) +
  theme(axis.text = element_text(color = "grey20", size = 10.2))+
  theme(plot.title = element_text(size = 18), plot.subtitle = element_text(size = 11, color = "grey20")) + 
  labs(x = NULL, y = NULL, 
       title ="No significant Difference of Order Priority by Customer Segment")+
  geom_text(data=df_for_ploting, aes(label=bar_text),
            position = position_stack(vjust=0.5))