R：查找元素重叠率 > 80% 的向量组

Question

我有一个向量列表，我想确定共享 80% 以上值的向量组。我希望使用组中包含的所有向量的名称来命名新组。我希望组的新元素是分组向量中的所有可能元素。我不确定这是否可行。

这是一个简单的例子：

a <- c(1,2,3,4,5,6,7,8,9,10,11,12,13,14)
b <- c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15)
c <- c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,16)
d <- c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17)
e <- c(1,2,3,4,5,6)
f <- c(18,19,20,21,22,23,24,25,26)
g <- c(18,19,20,21,22,23,24,25)
h <- c(18,19,20,21,22,23,24)
i <- c(18,19,1,2,3,4,5,6,7)

所以我认为我们会得到两组：

a-b-c-d-e-i
[1] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
f-g-h
[1] 18 19 20 21 22 23 24 25 26

向量中的值不必是数字，它们可以是字符串。让我知道是否还有其他需要指定的内容。谢谢！

Answer 1

假设向量在此答案末尾注释中显示的列表 V 中，计算邻接矩阵 adj，如果第 i 个和第 j 个向量至少有 80%共有元素，否则为 0。我们将此基于两个向量中较短者的分数。然后使用 igraph 包将adj 转换为图g 并计算其分量。

library(igraph)

overlap <- function(u, v) length(intersect(u, v)) / min(length(u), length(v)) > 0.8
adj <- sapply(V, function(u) sapply(V, overlap, u)) + 0
g <- graph_from_adjacency_matrix(adj)
memb <- components(g)$membership
memb
## a b c d e f g h i 
## 1 1 1 1 1 2 2 2 1 

s <- split(V, memb)
groups <- lapply(s, function(x) unique(unlist(x)))
names(groups) <- lapply(s, function(x) paste(names(x), collapse = "-"))
groups

给予：

$`a-b-c-d-e-i`
 [1]  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19

$`f-g-h`
[1] 18 19 20 21 22 23 24 25 26

我们还可以绘制图形：

plot(g)

注意：

V <- structure(list(a = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 
14), b = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15), 
    c = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 16), 
    d = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 
    16, 17), e = c(1, 2, 3, 4, 5, 6), f = c(18, 19, 20, 21, 22, 
    23, 24, 25, 26), g = c(18, 19, 20, 21, 22, 23, 24, 25), h = c(18, 
    19, 20, 21, 22, 23, 24), i = c(18, 19, 1, 2, 3, 4, 5, 6, 
    7)), .Names = c("a", "b", "c", "d", "e", "f", "g", "h", "i"
))