如何在 pca biplot 中显示观察结果？答案

【问题标题】：How to display observations in pca biplot?如何在 pca biplot 中显示观察结果？
【发布时间】：2021-12-27 09:49:32
【问题描述】：

我希望使用 factoextra 包在主成分分析双图上显示观察行名称。

library(factoextra)
## organize dataset to obtain unique rownames
iris$id <- rownames(iris)
iris$id <- paste(iris$Species, iris$id, sep="_")
rownames(iris) <- iris$id
iris <- iris[,1:4]

## Run PCA
pca <- prcomp(iris, center=T, scale. = T)

## On this plot observation names are displayed
fviz_pca_ind(pca,
         col.ind = "cos2", # Color by the quality of representation
         gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
         repel = TRUE
)

)

## Is there any way to display the same row names in this plot? (instead of points)
fviz_pca_biplot(pca,
            col.var = "#2E9FDF", # Variables color
            col.ind = "#696969",  # Individuals color
            label = 'var',
            repel = TRUE
)

提前致谢

【问题讨论】：

标签： r ggplot2 pca biplot factoextra

【解决方案1】：

您可以指定 geom.ind = "txt"

fviz_pca_biplot(pca,geom.ind="text",labelsize=2,
col.var = "#2E9FDF",col.ind = "#696969")

并尝试添加ggrepel：

options(ggrepel.max.overlaps = 20)
fviz_pca_biplot(pca,geom.ind="text",labelsize=2,
col.var = "#2E9FDF",col.ind = "#696969",repel=TRUE)

【讨论】：