Taipower-filter-top10-company

library(reshape2) 
library(dplyr)
library(vegan)
library(ggplot2)
library(ggdendro)

Taipower <- read.csv(("https://johnsonhsieh.github.io/DSC2016-R/data/Taipower_top100.csv"),fileEncoding = "big5")

aop = Taipower$attr_of_procurement
T = mutate(Taipower, category = substr(aop,1,5), category_detail = unlist(lapply(strsplit(as.character(aop), split = ">"),"[",2)))


sum_by_procurement = group_by(T, category) %>% 
  summarise(sum(as.numeric(total_tender_awarding_value), na.rm=TRUE))

avg_by_procurement = group_by(T, category) %>% 
  summarise(mean(as.numeric(total_tender_awarding_value), na.rm=TRUE))

T1 = filter(T, category == "<財物類>") %>% 
  group_by(tenderer_name) %>% 
  summarise(sum_of_value = sum(as.numeric(total_tender_awarding_value), na.rm=TRUE)) %>%
  arrange(desc(sum_of_value)) %>% head(10)

T2 = filter(T, category == "<勞務類>") %>% 
  group_by(tenderer_name) %>% 
  summarise(sum_of_value = sum(as.numeric(total_tender_awarding_value), na.rm=TRUE)) %>%
  arrange(desc(sum_of_value)) %>% 
  head(10)

T3 = filter(T, category == "<工程類>") %>% 
  group_by(tenderer_name) %>% 
  summarise(sum_of_value = sum(as.numeric(total_tender_awarding_value), na.rm=TRUE)) %>%
  arrange(desc(sum_of_value)) %>% 
  head(10)

T4 = group_by(T, category_detail, category) %>% 
  summarise(sum_of_value = sum(as.numeric(total_tender_awarding_value), na.rm=TRUE)) %>% 
  head(30)

T5 = group_by(T, category_detail, category, tenderer_name) %>% 
  summarise(sum_of_value = sum(as.numeric(total_tender_awarding_value), na.rm=TRUE))

group_by(T5,tenderer_name,category_detail) %>% summarise(Value = max(sum_of_value))
# T (data), sum_by_procurement, avg_by_procurement
# T$category, T$category_detail
# T1(財務 top10 公司), T2(勞務 top10 公司), T3(工程 top10 公司)
# T4(各子類別金額加總)
# T5(各子類別 top1 公司)