有没有办法在 R 的三元图上使用不重叠的数据点标签？

Question

我正在尝试复制我在 FiveThirtyEight 在https://fivethirtyeight.com/features/how-cable-news-reacted-to-the-cohen-hearing/ 上找到的图表。这显示了一个三元图，其中 3 轴中单词的位置显示了相应网络引用的比例。

我目前正在使用 R、ggplot2，更重要的是 ggtern（我广泛用于三元图）。但是，我从来没有找到一种方法使点上的数据标签不重叠。我一直希望 ggtern 能与 ggrepel 交互，但遗憾的是它没有（据我所知）。 有什么方法可以强制它们进行交互，或者找到其他方法吗？

为了清楚我所追求的，链接中显示的图表：

我的图表示例，其中单词重叠且看起来很糟糕：

编辑 创建我的可怕图表的代码：

    data <- data.frame(word = c("A","random","set","of","words","that","can","hopefully","help","someone","solve","my","issue","of","overlapping","labels","and","make","my","chart","readable","and","a","good","visualization"),
               axis1 = sample(1:100),
               axis2 = sample(1:100),
               axis3 = sample(1:100))

    ggtern(data = data,
        aes(x = axis1, y = axis2, z = axis3, colour = word, label = word)) +
      geom_point(size = 1) +
      geom_text()

Answer 1

好的，所以你想要ggrepel 包中的功能。虽然 ggrepel 在这里不起作用，但您可以使用 position_nudge_tern 和 check_overlap：

word = c("A","random","set","of","words","that","can","hopefully","help","someone","solve","my","issue","of","overlapping","labels","and","make","my","chart","readable","and","a","good","visualization")
col = c("red", "blue", "green", "red", "blue", "green","red", "blue", "green", "red", "blue", "green","red", "blue", "green", "red", "blue", "green","red", "blue", "green", "red", "blue", "green","red")

n  = 25   #Number of Data Points
nv = 0.1  #Vertical Adjustment
pn = position_nudge_tern(y=nv,x=-nv/2,z=-nv/2)

data <- data.frame(x = sample(1:25),
               y = sample(1:25),
               z = sample(1:25), 
               label=word)

ggtern(data = data, aes(x = x, y = y, z = z, colour = col, label = word)) +
  geom_point(size = 1) +
  theme_nomask() + #Allow Labels to Spool Over Edges
  geom_text(position=pn,aes(label=word),check_overlap=T, size=5)

哪个会给你不重叠的标签：