[Day-25] R语言 - 分群应用(五) 分群预测 - 资料清洗 ( data cleaning in R.Studio )

您的订阅是我制作影片的动力
订阅点这里~

影片程序码(延续昨天)

#步骤一: na补值&相异点or离群值侦测
library(naniar)#前置1: chk NA
library(clustMixType)
any_na(data) 
data_num <- data[,c(6,13:24)]
data_num <- as.data.frame(scale(data_num)) #前置2: standard norm 标准化
data_cla <- data[,c(2:5,7:12)]
data_cla <- lapply(data_cla, factor)
data_use <- cbind(data_num,data_cla)
kp <- kproto(data_use, 10)
kp_out <- as.data.frame(kp$cluster) #大好
final <- cbind(data,kp_out)

out <- final %>% 
  group_by(`kp$cluster`) %>% 
  summarise(笔数=n())

done <- final %>% 
  subset(`kp$cluster`!=10&`kp$cluster`!=6) 

若内容有误，还请留言指正，谢谢您的指教