这应该是一个很容易的问题,但我真的不知道如何做到这一点.我有,比如说,图中的图表:
每个节点都有一个日期,我想找到具有最大出度的节点,只有那些在中位数之前有日期的节点.
我试过这个:
library(igraph)
nodes <- data.frame(name=c("a", "c", "d", "e", "f", "g", "i", "j", "k"),
date = c(27, 13, 0, 18, 0, 8, 44, 26, 22))
relations <- data.frame(from=c("d", "d", "f", "f","f", "g","g","g","c","c", "e"),
to=c("i", "f","d","c","g","k","a","c","a", "e","j"))
ggg <- graph.data.frame(relations, directed=TRUE, vertices=nodes)
V(ggg)$label <- V(ggg)$name
plot(ggg, layout = layout.fruchterman.reingold.grid, edge.curved=FALSE,
edge.arrow.size=0.2,edge.arrow.width=0.4)
V(ggg)$label <- V(ggg)$date
plot(ggg, layout = layout.fruchterman.reingold.grid, edge.curved=FALSE,
edge.arrow.size=0.2,edge.arrow.width=0.4)
median_delay <- median(V(ggg)$date)
vert_before_median <- V(ggg)[ V(ggg)$date <= median_delay ]
wnodes <- V(ggg)$name[ degree(ggg,v=vert_before_median,mode="out")==max(degree(ggg,v=vert_before_median,mode="out")) ]
一切似乎都可以到这里:
> degree(ggg,v=vert_before_median,mode="out")==max(degree(ggg,v=vert_before_median,mode="out"))
c d e f g
FALSE FALSE FALSE TRUE TRUE
但是,当我想要保存满足此属性的节点时,我遇到了麻烦.我认为wnodes应该包含节点“f”和“g”
> wnodes
[1] "e" "f" "k"
似乎我在尝试从图中选择顶点的方式中遗漏了一些东西.我试过了,但它仍然是不对的:
> V(ggg)[which( degree(ggg,v=vert_before_median,mode="out")==max(degree(ggg,v=vert_before_median,mode="out")) )]
Vertex sequence:
[1] "e" "f"
任何的想法?
最佳答案 子集vert_before_median而不是所有顶点,你应该得到你所追求的,
wnodes <- vert_before_median[ degree(ggg,v=vert_before_median,mode="out")==max(degree(ggg,v=vert_before_median,mode="out")) ]
在上面的代码中,当您使用长度为9的所有顶点时,R会自动循环遍历长度为5的逻辑向量(FALSE,FALSE,FALSE,TRUE,TRUE).