Question

我有一个变量之间具有相关值的df：

input <- data.frame(Var1 = c("A","B","A","D","G", "H", "I"), Var2 = 
c("B","C","E","F", "F", "J", "K"), Corr_Value = c(1,1,1,0.7,0.7, 1,1), 
stringsAsFactors = F)

如图所示（A = B），（A = E）和（B = C）

我想获得一个类似于“输出”的df，其中所有变量关系都在一行中（A = B = C = E）

output <- data.frame(Var1 = c("A","D", "H", "I"),
Var2 = c("B","F", "J", "K"),
Var3 = c("C","G", "NA", "NA"),
Var4 = c("E", "NA", "NA", "NA"),
Corr_Value = c(1,0.7,1,1))

我该怎么做？

Answer 1

我们可以使用igraph来获得预期的输出

library(igraph)
g1 <- graph.data.frame(input[-3], directed = TRUE)
cl <- clusters(g1)
lst <- cluster_edge_betweenness(g1)
lst1 <- lst[1:4]
attr(lst, "class") <- NULL


out <- setNames(do.call(rbind.data.frame, lapply(lst1,
           `length<-`, max(lengths(lst1)))), paste0("Var", 1:4))
out$Corr_value <- input$Corr_Value[match(mapply(`[`, lst1, cl$csize), input$Var2)]
out
#  Var1 Var2 Var3 Var4 Corr_value
#1    A    B    C    E        1.0
#2    D    G    F <NA>        0.7
#3    H    J <NA> <NA>        1.0
#4    I    K <NA> <NA>        1.0

连续将具有相同相关值的变量合并

1 个答案: